ANDREW MORRISON (ed.) ed.)

Researching ICTs in Context

InterMedia Report 3/2002 University of Oslo

© InterMedia and the authors 2002 All rights reserved InterMedia University of Oslo Box 1161 Blindern 0317 Oslo Norway Report 3/2002 ISBN 82-8064-003-7 ISSN 1502-7198 Produced in co-operation with Unipub forlag The report may be ordered from www.gnist.no Cover: Alien Design AS, art work by Synne Skjulstad Printed by GCSM AS, Oslo 2002

Contents 1. 2.

Introduction

Andrew Morrison

Doing innovative ICT-research: methodological challenges in leveraging the best of three worlds

13

Public places – public activities? Methodological approaches and ethical dilemmas in research on computer-mediated communication contexts

33

Helge Godø

3.

Janne C.H. Bromseth

4.

1

Locating the Internet: studying ICT is passé in the culture of real virtuality 63

Trond Arne Undheim

5.

Beyond the enigmatic utopia: researching facts and failures in ICT projects

93

Experiments along the bazaar-route: the importance of user-producer dialogue in shaping new media technology

121

ICTs in contexts: reporting research from researching researchers' reports

141

Jarle Brosveet

6.

Per Hetland

7.

Beate Elvebakk

i

8.

The multi-dimensional stories of the gendered users of ICT

161

Seduced by numbers?

185

Researching performance, performing research: dance, multimedia and learning

211

Blogging thoughts: personal publication as an online research tool

249

Hilde Corneliussen

9. 10.

Helen Jøsok Gansmo

Synne Skjulstad, Andrew Morrison & Albertine Aaberge

11.

Torill Mortensen & Jill Walker

ii

Preface In the past decade, inquiry into Information Communication Technologies (ICTs) has been a major area of investment in research in Norway. In this respect, the Social and Cultural Presuppositions for Information and Communication Technologies (SKIKT) programme is one of the core areas funded by The Research Council of Norway. The SKIKT programme comes to a close at the end of 2002 and many of the SKIKT-funded researchers are beginning to consolidate a range of short and long term research projects. Insights from the various projects will be summarised for a wider public when the programme is formally completed. This edited collection, Researching ICTs in Context, is a demonstrable outcome from the SKIKT programme. Researching ICTs in Context is an attempt to reflect on some of the troubling and inspiring issues surrounding the study of new, digital media, systems and their social and cultural uses and interpretations. The book indicates some of the variety of research carried out by SKIKT-funded researchers. This variety is augmented by the research conference of the same name, held in Olso on 8 April 2002, for which this print text was compiled. The chapters from the book are also available on the Web and will be extended with material from the conference. This will take the form of slides from the presentations by the authors of the chapters. Invited guests and commentators, as well as other SKIKT supported researchers will contribute to the discussion of the papers. Notes, extracts and images from these discussions will be linked with the conference website as a way of also mediating research electronically. In preparation for further research on social and cultural aspects of ICTs, the book, conference and website, discuss how research might be performed in electronically mediated environments and communication. This relates to methodologies, but also to the uses of ICT when researching ICTs.

iii

As the host of one of the two strategic university programmes under SKIKT, InterMedia at the University of Oslo is pleased to be able to support this publication and conference. We do so in acknowledgement of the many years of hard work which have gone into the SKIKT programme.

Knut Lundby Professor Director of InterMedia, University of Oslo Member of the SKIKT board Oslo, 19 March 2002. www.intermedia/konferanser/skikt-02/skikt-research-conferance.html

iv

Contributors Albertine Aaberge, researcher, InterMedia, University of Oslo http://www.intermedia.uio.no/ansatte/ Janne Bromseth, doctoral student and researcher, Institute for Interdisciplinary Cultural Studies, Norwegian University of Science and Technology (NTNU), Trondheim http://www.hf.ntnu.no/itk/kv_bromseth/bromseth-index.htm Jarle Brosveet, researcher, Institute for Interdisciplinary Cultural Studies, Norwegian University of Science and Technology (NTNU), Trondheim http://www.hf.ntnu.no/itk/sts_brosveet/brosveet-index.htm Hilde Corneliussen, doctoral student and researcher, Department of Humanistic Informatics, University of Bergen http://www.uib.no/People/hhihc/ Beate Elvebak, doctoral student and researcher, Centre for Technology, Innovation and Culture (TIK), University of Oslo http://www.tik.uio.no/ansatte.html Helen Jøsok Gansmo, doctoral student and researcher, Centre for Technology and Society, Norwegian University of Science and Technology (NTNU), Trondheim http://www.hf.ntnu.no/itk/sts_gansmo/

v

Prof. Helge Godø, Head of the SKIKT board; Senior Researcher, Norwegian Institute for Studies in Research and Higher Education http://www.nifu.no/ansatte/helge.html Dr. Per Hetland, Director of Research, Eastern Norway Research Institute http://www.ostforsk.no/ansatte/visansatt.cfm?Ansattkode=47 Dr. Andrew Morrison, researcher and lecturer, InterMedia, University of Oslo http://www.intermedia.uio.no/ansatte/ Torill Mortensen, lecturer and doctoral student, Media Studies, Høgskule in Volda http://www.hivolda.no/amf/tilsette/mortensen/index.htm Synne Skjulstad, researcher, InterMedia, University of Oslo. http://www.intermedia.uio.no/ansatte/ Trond Arne Undheim, doctoral student and researcher, Centre for Technology and Society, Norwegian University of Science and Technology (NTNU), Trondheim http://www.hf.ntnu.no/itk/sts_undheim/ Jill Walker, doctoral student and researcher, Department of Humanistic Informatics, University of Bergen http://cmc.uib.no/jill/

vi

1 Introduction Andrew Morrison

On genres, audiences & media Research on ICTs The genre of the essay, as Hesse (1998) reminds us, may be traced back to the artistry of Montaigne. At that time, expository discourse was first a matter of composition by quill and ink, blotter, messy drafting and recopying. A genre of an information communication technology (ICT), the essay was invaluable in the positing of argument and the demarcation of its disposition. It enabled ideas to be circulated, albeit amongst the literate few of its time. However, as we know, another information communications technology, the printing press, was to change our notions of rights and access to the written word. Today, book collections debate Global Literacies and the World Wide Web (Hawisher & Selfe 2000), and online journals invite the electroniconly publication of research. Yet, after several decades of intense interest in ICTs, the technical and communicative potential of digitally mediated communication would appear to continue pose a number of challenges to our approaches, methods and practices as researchers. This collection of chapters draws on a range of research carried out in Norway from the mid-1990s, funded by the SKIKT programme (see Preface for details). SKIKT has supported many different projects, ones which have varied in length, complexity and scale. These projects have repeatedly pointed to the importance of seeing ICTs in context, that is in defining their situatedness either informationally, technologically or communicatively. Valuable research has been carried out by SKIKT 1

2

Researching ICTs in Context

funded researchers in such inquiry. On occasions — in the enactment of projects, at seminars and conferences and in research publications — elements from these three strands of ICT research have intersected to produce interdisciplinary results and knowledge of research processes. The research has been able to draw on the knowledge of Social Science and Technology Studies (SSTS), for example through the Centre for Technology, Innovation and Culture (TIK) at the University of Oslo. It has also been able to deploy interdisciplinary research methods and approaches to ICTs as culturally and as technically constructed, such as has been apparent in the work of many colleagues at the Norwegian University of Science and Technology (NTNU), Trondheim.

Problems & tools: a mix of the new & the old Information Communication Technology (ICT) has been a useful, if portmanteau-like, term for a cluster of elements concerning the arrival, spread and insidious presence of digital media, communication environments, and our social and cultural shaping, making and consumption of them. From seeing the computer as a medium (Andersen et al. 1994), drawing on semiotics and media studies, we have increasingly seen digital media as part of the changing cultural fabric of electronically mediated communication. As social actors and agents in computer mediated environments and communication, we have needed to bring social theory into technology design (Berg 2000). Design studies itself is now involved in iterative redesign. This has become clearly apparent in the past few years as Web design and graphics have been transformed by new computational software and dynamic databases, ones generated on the basis of users' actions, requests and interests. While the Web has grown faster than science fiction writers have been able to terraform new plots, a medley of hand held communication devices, and most recently merged telephony and PDAs, have taken us off desktops and into information architectures of different scales, mobility and exchange. In the 1980s and 1990s, e-mail and online chat developed idiolets and terms of their own. Now, the more recent explosion in SMS is about to be hoisted into the realm of almost-3G mobility. As our definitions and understandings of ICTs face another test with the promise of a multimediational and mobile 'revolution', we ought to be reminded of the many occasions when technology determinists have brazenly directed future scenarios and practices.

Introduction

We are weary of hearing that our access to, and use of, information depends on better and faster tools. So too are we more than a little jaded by the promises of both cyber community and e-commerce. The global economic downturn and dot-gone malaise have reigned in a rampant growth in digital commmunications and unrealistic expectations that the continued growth of a 'virtual' economy can be sustained. Earlier utopian projections of cybercommunity and along with them an intense creativity linked with the generation of new ideas, work practices and theories, are neverthesless themselves partial prisoners to commercial concerns and to corporate interests. Previous notions of community and of the Internet as a public sphere of sorts have been shown to be more complex communicative spaces, in terms of discourse and communication encounters than we had imagined, and in the ongoing interplay between users and tools. The crossing of personal, professional and public boundaries has become more complex than was naturally anticipated in earlier writings on cybersociety.

Digital media, communication & power Increasingly, relations between ICT-rich and less-wired communities and countries have moved from rhetoric into pragmatic projects and research. ICTs are now a part of 'development communication' just as they drive much of the R&D sector of their own development. But, then, in 2001 power shortages in California challenged the very assumption of a technology-safe silicon industry, and along with it givens about the domestication of technologies in the more affluent parts of the world. These surges and dips have been seen in global sales of computer hardware and software in the past few years, perhaps most markedly since September 2001. The world's smartest bombs have been pitted against some of the world's most rugged terrain in Afghanistan, yet the complexity of the material world and our human selves has shown that, despite the agency of machines, information and technology are part of our social and cultural experience and legacy. We may be beginning to ascribe agency to machines which act, just as we may ascribe roles to characters in the telling of failed ICT style research projects (Latour 1996). In these relationships between information structures, machine processed codes and our enculturated relations to them — as designers, learners or consumers — we have come

3

4

Researching ICTs in Context

to appreciate the value of contextualised interpretations and their variation. However, to carry through these interpretations in the face of interested players in the ICT market place is at times a delicate and difficult endeavour for researchers. At one level, as social scientists and humanists we need to know more about informatics, data processing and online publication and communication at the level of tools and techne. At another, cultural level, we need deeper knowledge of a variety of digitally mediated commmunication events and practices. At yet another level, in order to be able to account for and interpret emerging hybrid media types, genres and communicative practices, we need more detailed first-hand knowledge of them.

On the chapters The value of diversity This collection of ten papers by thirteen authors in total contains a variety of research on ICTs. Some of the chapters are by established senior researchers in a position to review the field, some are by doctoral students inquiring into emerging communicative pattern and genres. We have also attempted to include material from Trondheim, Bergen and Oslo. Hopefully, this variety indicates the diversity and depth of inquiry concerning ICT-related research in Norway. The collection is not presented as a bird's eye view of such research; the chapters are offered as a mixture of consolidated research, reflections on previous projects, work recently completed and research in progress. Together, these are the contexts within which we work and learn from one another. Increasingly, these are interdisciplinary contexts. As a whole, the authors argue that we view ICTs reflexively: ICTs are part of the dynamic of technical and human interaction in which our own roles as researchers are increasingly being technologically mediated. How are we as humanists and as social science researchers (and teachers) to relate to the ways in which our research environments and contexts may be changing? How too are we to respond to new, different demands placed on our research practices and competencies? Ethical and procedural questions may arise from specific ICT contexts. The social and cultural formations of different ICT contexts may

Introduction

demand ameliorated research approaches, the selection of different software, and perhaps suggest we seek out alternative and less bounded paths to interpretation. The chapters in this collection approach these issues from different perspectives; the authors refer to various contexts for conducting and interpreting ICT-related research, a summary of which now follows.

In summary In the first chapter, 'Doing innovative ICT-research: methodological challenges in leveraging the best of three worlds', Helge Godø discusses some of the methodological challenges involved in three modes of representing knowledge production systems and methods. Godø demarcates these as an explanations oriented world, a solutions oriented world, and a futures oriented world. The established disciplinary boundaries of university and academic centred knowledge production allows us to concentrate on the why of knowledge making. In contrast to the formalism and often hierarchical configuration of such a knowledge institution, in an solutions oriented world, the context for knowledge production moves from the why to the “how&now” characteristic of the R&D sector. Godø refers to the third mode, the futures oriented world, as being concerned with the ongoing development of ICTs along with matters of agenda setting, the generation of policies and understanding of the political dimensions of knowledge production more generally. In his chapter, Godø argues that the first two of these knowledge modes are interlinked but that their lineaege is often one of patchy relations. The academy tends not to be well linked with the other two knowledge making arenas. A host of ethical and methodological questions may arise when we begin to research public, virtual spaces. In her chapter, 'Public places – public activities?', Janne Bromseth discusses some of these questions with respect to the Internet. She outlines how problems may arise about the use of public communication in the electronic domain, especially when it is technically possible to record, present, distribute and manipulate it. She asks how far researchers should go in using such material without the consent of those who produce it and for whom it has other intentions. Referring to electronic discussion groups, Bromseth, challenges us to reconsider the relationships between our practiced re-

5

6

Researching ICTs in Context

search methodologies and those both possible and in need of scrutiny in computer mediated communication (CMC). She also asks what practical challenges different genres of CMC may pose for us practically. What relations between practice, ethics and research might we consider in rethinking the public and private spheres of communication in digitally mediated environments. 'Real virtuality' is at the centre of the imagined community and imagined communication which is narrated and analysed in the chapter by Trond Arne Undheim. Entitled 'Locating the Internet: studying ICT is passé in the culture of real virtuality', Undheim narrates a fictional lecture in which a teaching assistant attempts to locate the Internet, as it were. This he does with reference to communication by many luminaries of ICT research. Castells, for example, is inscribed through his avatar. Undheim deploys this rhetoric as a means of contextualising the virtuality of electronic communication. He invites us to join him in this 'interplay' as part of the 'naturalisation' of the Internet in northern nations. However, he does so in part to discuss the some of the assumptions concerning the domestication of ICTs, mobility and nomadic work. Undheim is concerned with how we might engage in a double dynamic in which we differentiate between the making of on and offline spaces as well as their transgression via media, artefacts and people. 'Beyond the enigmatic utopia' is the phrase Jarle Brosveet selects to introduce his chapter on problems of evaluating ICT' projects. Brosveet uses three cases as illustrations of his own experiences as a researcher who is involved in trying to establish facts in ICT projects. The three cases show just how varied, interesting and frustrating this can be. Tellingly, Brosveet reminds us all of our own roles as researchers in other people's worlds, especially the extent to which we need to be constantly aware of the constructedness of discourses about ICT research. He argues that we often find a mismatch between the facts of projects and the accounts given of these projects by their different stakeholders. In distinguishing between the reporting on projects and the seemingly wishful review of some project members, Brosveet indicates just how difficult it is to conduct research in the contexts of projects' own histories, records and recollections, all the more difficult when these have negative results or outcomes. His three cases present this as a nuanced set of scenarios and from the occasional chill in the air one feels in the scenarios Brosveet presents,

Introduction

Per Hetland shifts us metaphorically and contextually. In his 'Experiments along the bazaar-route: the importance of user-producer dialogue in shaping new media technology', Hetland leaves the beaten track of the 'direct' and 'middleman' routes to science and technology communication. Instead, he suggests we follow 'the bazaar-route' as an expression for a complex, and 'hot', communication situation, one in which users are present. Hetland argues that in the shaping of new, media technologies, what is important is the way in which dialogue over what constitutes relevant knowledge takes place between the various actors involved. He exemplifies this with reference to the social shaping of communications technologies designed for and used by people with partial and restricted sight. In his analysis, he shows how user values may be diffuse and that users may produce different outcomes to those intended by designers of communications technologies. He suggests that demonstration experiments provide evidence of how interactions between these actors may be recast. In her chapter, 'ICTs in contexts: reporting research from researching researchers' reports', Beate Elvebakk takes us into the interesting problematics of how to research and to present accounts of uses of ICTs by professionals. Adopting the interview as her primary research instrument, Elvebakk highlights two problems for the researcher of ICTs. First, in their professional fields, here Chemistry, researchers have a tacit knowledge of their uses of ICTs, and this is complicated by recollection of past uses. Second, reporting on the uses of ICTs in research may itself be understood as a form of professional interpretation. Elvebakk argues that for the researcher studying such professional uses of ICTs, several of the leading studies of commuication and digital technologies are unsatisfactory. She offers several suggestions as to how to work though a 'precarious methodological situation' if we are to be able to 'approach truth' in our inquiries into ICTs. ICTs are often seen along gendered lines, with questions of access, uses, training and the gendered character of much of the IT industry. In her chapter, Hilde Corneliussen asks, 'How can we understand the gendered users of ICT?' In previous research she found that woman may downplay their knowledge of computers. To investigate this further she embarked on a study of students of computing. Her research methods included interviews, observation and communication via e-mail. In accounting for and interpreting the data from these combined approaches, Corneli-

7

8

Researching ICTs in Context

ussen links gender theory and discourse theory. In summary, she finds discourses of computing as still being clearly engendered. This gendered interest is continued in the next chapter by Helen Jøsok Gansmo, who asks, given the variety and volume of data on gender and computing, whether we are 'Seduced by numbers?' She presents material from a larger study of how policy makers and politicians do and do not make use of social science research in their bid to develop universal access and competence in schools. In this decisionmaking process, Gansmo demonstrates that qualitative studies tend to be overshadowed by reference to quantitative data. She argues that it is vital that we pay close attention to how girls do and do not approach, use and talk about their experience and needs when it comes to computing. In attempting to debate the context of ICT related research, Gansmo asks that we find ways to communicate qualitative research results on girls and computing so that their scientific value may be appreciated by decision-makers. 'Researching performance, performing research: dance, multimedia and learning' is the title of the chapter by Synne Skjulstad, Andrew Morrison and Albertine Aaberge. Relatively little research has been published on the intersection of dance and ICTs. The chapter provides a case based presentation and discussion of the links between digital media and experimental dance. In conjunction with a choreographer and dance educator, a media design and research team worked with students in developing a dance performance. The chapter presents this in the frame of performance studies, arguing that this field may be usefully included in the interdisciplinary approaches needed to make digital media and to analyse it. The authors take the blend of hermeneutic and qualitative inquiry from 'researching performance' across to the 'performing of research'. They argue that the documention and interpretation of ICT research may now be electronically linked, especially via the Web, suggesting that we include the study of an electronic research rhetoric as part of contextualising ICTs. The final chapter, by Torill Mortensen and Jill Walker, is called 'Blogging thoughts: personal publication as an online research tool'. Mortensen and Walker report on their own uses of weblogs or blogs, as a means to keeping diaries in which they process ideas about their research. Blogs are continuous and emerging texts; they are publically accessible; they provide an ongoing version of the research process.

Introduction

Blogs may be searchable but they are still largely untheorised. The researchers show how blogging has influenced their own research and they relate this as part of process-related communication and research. The authors make extensive reference to the rapidly emerging phenomenon of blogs through their detailed knowledge of them. Referring to their own experiences as 'bloggers', they suggest that blogs — texts which occur only through their electronic intertextuality — may be seen as a new form of research tool for inquiring into ICTs and contexts.

Research in a digital age Challenges to research methodologies? The varied concerns and approaches in these chapters suggest some, if not all, of the issues present in ongoing research into information, communication and technology in the humanities and social sciences in Norway. The title of this book and its accompanying conference, Researching ICTs in Context, indicates that our research community, diverse and interdisciplinary as it may be, has a common interest in the study and interpretation of ICTs as part of contextualised inquiry. Are our learned and practised methods and approaches adequate preparation for new phenomena, or for expected happenstances in new, virtual cladding? While the audio-casette recorder greatly eased the transcription of face-to-face interviews, and helped us to pay attention to the people we were interviewing in different ways, in what ways might the seemingly possible merger of media types in online documents and communication challenge our research skills and strategies? When, for example, audio files may be accessed and annotated from within research texts, and where audio may be used in formal academic communication as a mode of expository discourse, what does this imply for our own changing literacies as researchers? Our well-worn, indeed valuable, means of conducting and communicating our research ought to continue to be exercised. But, how might they be transformed in electronically mediated research and its online publication? 'Writing in the hivemind' (Byrd & Owen 1998) is one experiment at collaborative peer based academic communication among teachers of writing. How might we extend this to investigate a multi-

9

10

Researching ICTs in Context

modal electronic research discourse on and in ICTs (cf. Welch 1999)? Two often cited works on carrying out Internet related research (Jones 1999, Mann & Stewart 2000) point to the need to tackle new, remote, multiply constructed communicative and research domains. What, though, of our own identities, theories and methodologies in communicating about online and digitally mediated information and communication? How are we, for example, to learn how to develop the new, visual essay unless we develop a digitally informed hermeneutics of seeing (e.g. Davey 1999), one co-joined with that of 'composition'? What role, too, might digital aesthetics play in this new multi-modal research stylistics (see Cubitt 1998).

Communicating via screenspaces Publishing these chapters in book form is one part of a wider strategy to mediate the SKIKT-funded research to a wider research and learning community. The authors of the chapters will present them for discussion at the conference, held at InterMedia, the University of Oslo's new centre of 'new media and net-based learning'. The conference has two senior researchers and academics as guests. Prof, Sally Wyatt, University of Amsterdam, opens the conference, thematically, theoretically and with reference to changing research methodologies in the context of ICTs. Together with Prof. Lars Qvortrup, the University of Southern Denmark (Odense), Sally Wyatt comments on papers as a lead to discussion. Summarries of these comments and discussions appear on the conference website. The conference website also contains the full texts from this book, along with slides from the presenters. We have used this occasion as an attempt to also perform a shift away from our individual computer screens to a common, electroncially mediated communicative space. This is part of a larger attempt by InterMedia to mediate research online in experimental, heterotopic spaces (e.g. Galin & Latchaw 1998). These spaces are both investigations into the changing functions and character of genres (Bakhtin 1986) as well as 'theatres of experimentation' (e.g. Friedlander 1995) on ICTs s well as communicating about them (Usher 1997). The authors who have contributed to Researching ICTs in Context have provided us with a collection of what may at at times be provocative pieces about tackling procedural, ethical, theoretical and discursive aspects of re-

Introduction

search. We invite you to extend the communicative context of this book, by joining them and other readers via the conference website and its discussion forum (www.intermedia.uio.no).

References (All references and notes related to the World Wide Web are cited on 20 March 2002). Andersen, P., Holmqvist, B. & Jensen, J. 1993. (eds) The Computer as Medium. Cambridge: Cambridge University Press. Bakhtin, M. 1986. Speech Genres and Other Late Essays. Austin: University of Texas Press. Berg, M. 1998. 'The politics of technology; on bringing social theory into technological design', Science, Technology, & Human Values, Vol. 23, No. 4. 456-490. Byrd, D. & Owen, D. 1998. 'Writing in the hivemind'. In Taylor, R. & Ward, E. (eds) Literacy Theory in the Age of the Internet. New York: Columbia University Press. 47- 58. Cubitt, S. 1998. Digital Aesthetics. London: Sage. Davey, N. 1999. 'The hermeneutics of seeing'. In Haywood, I. & Sandywell, B. (eds) Interpreting Visual Culture: explorations in the hermeneutics of the visual. London: Routledge. 3-29. Friedlander L 1995 'Spaces of experience on designing multimedia applications' in Barrett E & Redmond M (eds) Contextual Media: multimedia and interpretation The MIT Press Cambridge MA 162-174 Galin, J. & Latchaw, J. 1998. 'Heterotopic spaces online: a new paradigm for academic scholarship and publication', Kairos Vol. 3, No. 1. At: At: http://english.ttu.edu/kairos/3.1/coverweb/galin/index.htm Hawisher, G. & Selfe, C. 2000. (eds) Global Literacies and the World Wide Web. London: Routledge. Hesse, D. 1999. 'Saving a place for essayistic literacy'. In Hawisher, G. & Selfe, C. (eds) Passions, Pedagogies and 21st Century Technologies. Logan: Utah State University Press. 34-48. Jones, S. 1999. (ed) Doing Internet Research: critical issues and methods for examining the Net. Thousand Oaks: Sage. Latour, B. 1996. Aramis or the Love of Technology. Cambridge MA: Harvard University Press.

11

12

Researching ICTs in Context

Mann, C. & Stewart, F. 2000. Internet Communication and Qualitative Research: a handbook for researching online. London: Sage. Usher, R. 1997. 'Telling a story about research and research as storytelling: postmodern approaches to social research'. In McKenzie, G., Powell, J. & Usher, R. (eds) Understanding Social Research. London: The Falmer Press. 27-41. Welch, K. 1999. Electric Rhetoric: classical rhetoric, oralism, and new literacy. Cambridge MA: The MIT Press.

2 Doing innovative ICT-research: methodological challenges in leveraging the best of three worlds Helge Godø

Introduction The main purpose of undertaking research is to create – in a wide sense - useful knowledge and innovations. Increasingly, more and more people are doing research and development, and, in particular, the ICTsector is characterised as R&D-intensive. Because of this, what research creates may have a significant impact on our future. Simultaneously, as this universe of ICT-related research is expanding, there is an increasing variety of research undertaken. In spite of this increasing complexity, one may claim that there are basically three modes and related methodologies for undertaking ICT-research. First, the type of research that has its centre in the academic world; secondly, the research affiliated with industry and the non-academic world, and, finally, a third, future-oriented mode. For various reasons that will be elaborated in the chapter, at present this third mode is not so strong, however, its past successes and achievements in the development of ICT deserve closer attention. In order to resurrect this third mode, one must be able to leverage strong assets from the two other modes of doing ICT research. The potential for increasing the usefulness of research related to social and cultural aspects of ICT seem favourable. However, this may require adjustments of research methodologies, which is why an eclectic approach is advocated.

13

14

Researching ICTs in Context

Eclecticism In approaching the topic of this chapter, it may be useful to elucidate the concept of eclecticism, because 'leveraging the best of three worlds' may be characterised as an eclectic ambition. The term eclectic is defined in the dictionary as 'selecting what appears to be best in various doctrines, methods, or styles'.1 In some social circles or contexts, this term is used as a derogatory characteristic, presumably for two reasons: first, 'eclectic' is equated with lack of originality (almost plagiarism), and, second, the reconfiguration of elements (the 'best') is an eviction of these out of their contexts in order to create something artificial. As implied in the judgment of something as being a 'total clash', the reconfiguration involved in an eclectic approach has been considered unsuccessful, because it violates inherent rules as to how various elements should be combined.2 Skeptics of the eclectic approach may point to numerous failures, such as Esperanto, i.e. the attempt to create a single, universal language, or the project of making a hybrid version of the airplane and the car, which was attempted in the carefree 1950s in the USA. Still, one may claim that a salient feature of most innovations is that they are eclectic; creating innovations demands an ability to recombine or reconfigure known elements in a new way – and by this, create a novelty, i.e. something new to the world, something that has not existed before. One may even stretch this point further: eclecticism is a basic principle in evolution; evolution is basically a process in which existing elements are selected and recombined so as to create something new. Indeed the term selection is in accordance with the etymological origin of 'eclectic', as this is derived for the Greek word 'eklektikos', which means to select. What differentiates biological evolution from sociocultural evolution is that in the latter, selection to a large extent may be associated with human choice, closely related to volition, will, design and planning, i.e. it rests on the ability to envisage something, that does not exist at present, may be created by recombining and redesigning known elements, which is essentially an eclectic approach. In bioevolution, chance and randomised, haphazard events select by default. Having justified that eclecticism in general may be considered as fertile and legitimate, I will now introduce the background for the topic of why eclecticism may be relevant for ICT research.

Doing innovative ICT-research

Structural changes in the ICT sector During the 1990s, a major structural change occurred as the ICT-industry, in particular the telecommunications industry, was deregulated, liberalised and, for this reason, radically restructured. This, as will be elaborated, had a profound impact on ICT-related research and development (R&D). The reasons for this are complex, however, a simple, sweeping explanation is: those with political power in the most powerful nations of the world (the largest OECD countries) wanted the 'markets to rule' in the ICT sector; they justly perceived the ICT-sector as dominated by monopolies ruled by technocrats and bureaucrats who were insensitive to customers' demands, charging outrageous prizes for low-quality services and products. The foundation for this was clearly ideological, as personified in the political attitudes of the US President Ronald Reagan and the UK PM Margaret Thatcher, who were in office during large parts of the 1980s.3 Both represented very powerful political forces and were simultaneously strong proponents for this movement, which emerged in the early 1980s. The shift towards market liberalism came just as ICT in general, but specifically the Internet and the mobile communication system GSM, 'took off' in earnest in the early 1990s, starting a process by which most industrial societies of the world have become ICT-saturated. In technological and economic terms, ICT was very successful. However, this success became associated with the policy shifts, i.e. that the dynamics of liberalism and free market rocketed ICT in its development and diffusion in societies, a case of 'success by association' as contrasted to 'guilt by association'. In all fairness, the shifts created a new climate and culture within the industry; because market competition was introduced, greater focus was set on profitability and business models. Inevitably, customers, that is those willing to pay for services and products, became important; for this reason focus was set on their demands and desires. Within a short period, the industry adopted aggressive marketing strategies that would have been unimaginable prior to the shift. However, there was no causal relationship between the ideological policy shift and the success of ICT – this was spurious. Nevertheless, the structural implications of the policy shift were profound, in particular for the research activities undertaken in the ICT-industry. As will be elaborated, the success of ICT in the 1990s were based on innovations that have developmental trajectories going back to R&D

15

16

Researching ICTs in Context

undertaken in the 1970s and 1980s, to the R&D-organisations of the industry and to a unique 'innovation regime' (Godoe 2000) in the ICTindustry, which was particularly strong in some parts of the telecommunication sector. The development of the highly successful GSM mobile communication system was initiated in 1979-1980 and took ten years to develop into a comprehensive, pan-European joint developmental effort involving a large number of R&D-facilities, to a cost of more than NOK 2 billion. No single nation or company had at that time the resources (money and people) needed for a project of this magnitude, however, by cooperating this was made possible. Just as important, however, the partners in this development (almost all the European telecommunications operators and equipment manufacturers), agreed to build, operate and provide GSM to the public, an agreement which was first signed as a Memorandum of Understanding in 1986. Thus, when GSM was launched, this occurred almost simultaneously in Europe. Probably, due to the success of this and similar joint R&D-efforts in the 1970s and 1980s, the ICT-industry in December 1998 established an organisation called 3GPP, which is an acronym for '3rd Generation Partnership Project' in order to act as a mechanism for harmonising various technical standards related to UMTS.4 The core of this initiative is ETSI, which was established in 1987, thus representing a direct continuation of the innovation regime associated with the ancien regime, but now in a different setting.5

Reorganising ICT-related R&D As most of the traditional ICT-companies were faced with prospects of struggling for survival in fiercely competitive markets, they started to reorganise themselves. BPR – the buzz-word for Business Process Reengineering – became the mantra for doing this. Subsequently, the companies were invaded by Anglo-American-speaking MBAs from expensive international consultancy firms (Andersen, McKinsey, Cap Gemeni, etc.) hired by the top management. In a short time, the old ICT-companies became partitioned, reorganised as autonomous business units that in theory were independent in terms of their operations. Simultaneously, what may be termed a 'market-oriented R&Dgovernance model' was introduced, i.e. the business units became responsible for acquiring and paying for the R&D they felt a need for.

Doing innovative ICT-research

The implications of this were dramatic because the work undertaken by the old R&D-units now became a commodity or service which they had to sell to the business units. In these, R&D-costs had to compete with other types of costs; if a business manager could avoid R&D-costs, he would be happy because this would improve the bottom line figures which became the ultimate standard for measuring business unit performance. The impact of this was a radical shift in the R&D undertaken by the ICT-companies. Focusing on the service and product portfolios of the business units, R&D became product development-oriented, responding to the immediate needs and aspirations of their clients. Needless to say, R&D related to product development is important and demands just as much talent, creativity and ingenuity as other types of R&D. However, the R&Dunits had to adjust and adapt themselves to new demands, reorient their attention, change their modus operandi, etc. Because elaboration of this point would be a sidetrack to the main theme of this paper, suffice with one example which illustrates one of the impacts: in 1991, the closely knit network of Nordic mobile communications researchers who had successfully cooperated for many years was closed down. This network had been pivotal in the development of the first generation mobile communication system NMT in the 1970s and played an important role in the development of the next generation mobile communications system (GSM) during the 1980s. The justification for closing down the network as twofold. The business units (who now had to pay for these activities) did not like the idea of their 'own' researchers collaborating with researchers from competitors; this might provide competitors with confidential business information. Second, this type of collaboration could be considered as market collusion detrimental to competition by anti-monopoly authorities in the EU, which could potentially back-fire on the firms. Thus, the shift proved to become incompatible with many of the basic institutions that motivated the successful innovation regime of the 1970s and 1980s. To cut a long story short, in refocusing and restructuring R&D, the longterm, high-risk, explorative reseach and international collaborative projects – what constituted the foundations of the innovation regimes that created the success of ICT in the 1990s – lost their strength just as they were proving their success – instead the success was attributed to the policy shifts. The traditional R&D became substituted by product and service developmentoriented R&D having a short time horizon, thus becoming market-driven.

17

18

Researching ICTs in Context

Simultaneously, some of the research that was abandoned or downsized, migrated to the universities, i.e. changed institutional affiliation and identity. This may be described as a dual, push-and-pull-process: Recognising the significance of ICT as a new knowledge domain, ICT gradually became an area of high priority at the universities.6 Coinciding with the shift described above, this provided opportunities for many researchers to migrate to the universities. In this, they became professors and teachers of knowledge areas that were new at many universities, ultimately starting the institutional saturation process involved in the construction of new academic disciplines. Thus, one may claim that during the 1980s and 1990s, in the world of ICT-research, not just one, but numerous babies were thrown out with the water as the ICT-industry, in particular the telecommunications industry, was deregulated, 'liberalised' and, ultimately, radically restructured. Luckily, some of these babies landed in the laps of universities; some babies were snatched mid-air by entrepreneurial people at these and similar institutions. In the post-2000, restructured ICT-world, doing long-term, exploratory, high-risk and non-commercial ICT-research has become a public responsibility. Universities and related public R&D institutes, agencies and research funding agencies (both national and international, such as the EU) have emerged in the institutional landscape as actors undertaking or promoting ICT research. In the meantime, the ICT-industry has shifted its main R&D-focus to 'time-to-market', i.e. short-term product and service development. As the dust is now settling down after these shake-ups, the time has come to think forward in terms of ICT-research. In the following sections this will be attempted, however, some theoretical and methodological aspects related to research and innovations need to be elucidated.

Different models of doing research and creating innovations In the scholarship and discourses on research, innovations and knowledge production during the past 15 years, debunking what is known as the linear innovation model has become a standard start-up ritual. The linear innovation model depicts knowledge production and innovations as initiated by basic research; basic research as the primordial source of knowledge provides input to applied research, which in turn creates an input for product and process development, this then creates an output for the final destination: production and service provision. The linear innovation model still has

Doing innovative ICT-research

strong and persistent advocates in university communities, in spite of overwhelming empirical evidence showing a different picture. The snowball that set off the avalanche in terms of criticism against the linear innovation model came in 1986, in an article written by Steven Kline and Nathan Rosenberg (1986). Pointing to hard empirical evidence, they claimed that the great majority of innovations are created in an intense interaction between markets and its users on the one side, and designers, or developers, of technology, products and services on the other side. Calling this the chain-linked innovation model, they depicted 'research' and 'science' as a cloud drifting above the innovation system, thus alluding to its detachment and irrelevance for innovations. In the text explaining this, they claimed that firms only undertake research when everything else fails, if they can afford it and if they think this will help. The model was highly successful in terms of popularity among policy makers; in fact OECD used this for a number of years, more or less as its official model of innovations (cf. OECD 1992). The interest for this lay in growing policy concerns for how to increase employment and competitiveness in the industry, this being particularly urgent in the ailing, old industrial economies of Europe. They justifiably claimed that the academic world, with their claims for the potential economic and social benefits of research, were not helpful, in spite of increased funding. Thus, the political climate was favorable for analyses and explanations on how to foster and create innovations, in particular a quest for understanding what kind of links exist between R&D and innovations and what kind of policy initiatives that should be taken in order to increase innovation output. Note should be taken that this took place simultaneously with the movement for economic liberalism described earlier. Among numerous interesting explanations, analyses and models that emerged in the ensuing debates, perhaps the idea of distinguishing knowledge production in society in two qualitatively different domains, termed 'Mode 1' and 'Mode 2', has created highest impact, perhaps more in Europe than elsewhere. This will be elaborated shortly, however, this way of analyzing knowledge production challenged established notions of institutional roles and identities, more so than the earlier critique of the linear innovation model. Apparently, these notions are more deeply entrenched in the tradition-bound, 'old' social institutions of Europe, compared with the instrumentality and pragmatism found elsewhere in the world (e.g. USA) in terms of knowledge production.

19

20

Researching ICTs in Context

The idea of distinguishing knowledge production in modern society in two domains, Mode 1 and Mode 2, was first published in a book (Gibbons et al. 1994)7 written by a group of science and technology researchers, of whom Michael Gibbons, as the first author, is the best known. Thus, for conversational convenience, many people identify this as 'Michael Gibbons' idea'. Indeed, one may justifiably characterise this as an 'idea', or, perhaps more accurately, as a conceptual framework articulating, in an essayistic style, with some (not many) attempts of empirical corroboration by pointing to a few cases, i.e. what critics term as 'anecdotal evidence'. However, in spite of this, the idea that knowledge production in modern society may be classified as consisting of two qualitatively distinct, yet equally significant modes has mobilised much interest and sympathy – and criticism. Not surprisingly, criticism came from those who adhere to, or identify themselves strongly with the linear innovation model and the implicit supremacy of the academic tradition as the primordial knowledge production institution of society. Those who are favorable to the ideas of Mode 1 and Mode 2 claim that this approach provides a more satisfactory framework for explaining modern societies in which the roles of knowledge, research and education are characterised by increasing distribution, complexity and significance. In part this is because of the development and diffusion of ICT and the transformations that many view as the emerging Information Society. For the purpose of discussing the methodological challenges posed by ICT research, the intention of this chapter is to attempt an extension of Michael Gibbons' idea with a new category of knowledge production, called Mode 3. In order to create a setting for this, a brief tour into Mode 1 and Mode 2, explaining its most salient features, will be undertaken. As will be shown, the main motivation for proposing a Mode 3 is that, although the dichotomy Mode 1 and Mode 2 may be fertile, these categories do not adequately fit the dynamics described earlier as a unique 'innovation regime' in ICT. This absence, and its implications, require a third category, hence Mode 3.

Mode 1 and Mode 2 The idea of Mode 1 and Mode 2 was first published in 1994, in the book, The New Production of Knowledge – The Dynamics of Science and Research in Contemporary Societies, with Michael Gibbons as its first author

Doing innovative ICT-research

(cf. Gibbons et al. 1994). The book points to numerous factors for justifying the idea, however, two of these need to be emphasised. The 'demographic' and institutional factor. After WWII, the number of people who have received higher education (university degrees at all levels) has increased rapidly, this also includes the number of people who have Ph.D.s, as depicted in the term 'mass education'. A large and increasing number of these graduates have entered into occupations related to knowledge production, at institutions located outside the traditional academic community, such as industrial R&D-facilities, consultancies, specialised governmental agencies and laboratories, technocracies, as experts and highly specialised professionals in private and semi-private organisations, media, NGOs, etc. These people constitute the main population of Mode 2. Simultaneously, the number and size of 'non-academic' (Mode 2) institutions involved in knowledge production have also increased.8 Increasingly, there are more Ph.D.s working outside the academic community – this in part because universities educate more Ph.D.s than they are able to absorb as employees, and partly because the outside world may offer better pay and more favorable career opportunities as these human resources are needed in their (Mode 2) knowledge production, thus confirming the proverb that 'talent follows money'. These trends have caused a radical redistribution of knowledge and knowledge production capability in society – the universities no longer command knowledge production hegemony. The distinction of basic vs. applied knowledge is unclear. The assumption underlying the linear innovation model implies a primordial status of the knowledge type called 'basic' or 'fundamental'. As the knowledge monopoly and hegemony of the universities have decreased (cf. the point above), the term 'basic research' has eroded as a distinct knowledge category, if this ever was a distinct category. The pragmatic reason for this is that classification of knowledge as 'basic' in terms of inherent properties of the knowledge is very difficult, making distinctions vague and contestable. However, the term may be operationalised in an institutional context, as in the OECD-statistics on R&D, where management and governance criteria determine categorisation of R&D, i.e. the term 'basic research' being most applicable to universities for these reasons. There are numerous cases of knowledge being 'reclassified' because of changes in institutional affiliation, such as the ICT-research undertaken at universities are now 'basic', whereas similar research undertaken in industrial laboratories are variously classified as 'applied research' or 'product development'. Similarly,

21

22

Researching ICTs in Context

numerous cases of new knowledge areas have risen, which do not fit into the dynamics of the linear innovation model. In the dichotomy Mode 1 vs. Mode 2, Gibbons (1994, 262) sees that: … in Mode 1 problems are set and solved in a context governed by, the largely academic, interests of a specific community. By contrast, Mode 2 knowledge is carried out in a context of application. Mode 1 is disciplinary while Mode 2 is transdisciplinary.

In elucidating the first of these modes, Gibbons points to the fact that disciplines such as material science, computer science and numerous disciplines related to engineering, had their genesis outside academia, due to lack of established knowledge, i.e. they had to start from scratch, evolving into new knowledge types, subsequently entering into academia from the 'outside', not as the result of any 'basic' research. Their origin was knowledge that had been created in the context of application. Thus a key concept in the ideas of Gibbons and his team is context, i.e. they emphasise that contextualisation of knowledge production is more significant than inherent properties of the knowledge itself. Using this as the basic explanatory assumption, they elaborate the difference between Mode 1 and Mode 2 in a number of salient factors, as will be explained below, and as shown schematically in Figure 1. Those who are strangers to this approach should be warned that this comparison of Mode 1 and Mode 2 is based on highly stylised characterisations, i.e. what is often termed ideal-typical, because these characterisations do not always fit neatly into the complex, murky world of empirical reality. Mode 1 Academic context Disciplinary boundaries Homogenity in perceptions Quality control and relevance defined by peers, i.e. within the discipline Hierarchical-static structure and organisation Internal accountability Academic freedom and 'quest for knowledge'

Mode 2 Context of application Multidisciplinarity Heterogenity in perceptions External principals decide on quality and relevance Ad-hoc organisations and flat structures External accountability Users and interests define the agendas

Figure 1: An outline of key characteristics of Mode 1 and Mode 2.

Doing innovative ICT-research

The implications of context in the knowledge creation and production is that in Mode 1, the academic context, knowledge production has a focus of attention on the discipline to which it belongs, i.e. the agendas that the discipline has decided as theirs. In contrast, in Mode 2, because attention is set on the applications, i.e. the agenda is defined by the challenge or problem that needs knowledge, this tends to favor a multitude of approaches in terms of Mode 1 discipline distinctions. Thus, the Mode 2 approach is transdisciplinary. Historically, transdisciplinary approaches haves been much more prevalent in technological disciplines, as indicated by the term polytechnic. Increasingly, this is also the case with social sciences and liberal arts to the extent these are applied in what Herbert Simon (1981) called the sciences of the artificial. The implication of this is that Mode 1's focus on disciplines creates homogeneity in the culture of its discipline, as evident when peers of a discipline are asked to make a judgment on a thesis or an examination: Independent of each other they tend to make almost identical judgment, making relevant almost similar criteria for this. Mode 2, in contrast, because of its compositeness, is more heterogeneous; numerous modes of understanding and interpretation, and strategies for creating a solution are offered because perceptions and past experiences are based on a much wider range of knowledge and problem solving strategies. In Mode 1, disciplines thrive and depend on stability and temporal continuity, thus it also tends to foster hierarchies, this stratification being governed by authority, which usually favors seniority and elderly men with similar socio-economic and cultural background. As typical of authority-oriented systems, those who are skillful in adjusting to, or manipulating authorities, wherever they may be (just think of Thomas Aquinas' position as an authority in the history of theology, perhaps the oldest Mode 1 discipline), have a greater probability of succeeding than those who are incapable of this. In Mode 2, as a contrast, the project or assignment is a typical organisational principle, i.e. an ad hoc, missionoriented organisation set up to solve a particular problem. For this reason, the organisations tend to be less stratified and more transient, thus the inevitable social status stratification processes do not promote themselves strongly. This is also related to the next aspect: accountability. In Mode 1, quality and relevance of knowledge is determined based on the judgment of peers. As the agendas are endogenously generated within the discipline, they are only accountable to themselves, however,

23

24

Researching ICTs in Context

this follows a top-down logic based on the authority hierarchy described above. In Mode 2, the accountability is external: the standards of quality and relevance are set and judged by outsiders, by principals or 'customers' who have vested interests in this type of knowledge. Anyone familiar with academic culture (Mode 1) must be aware of the tensions, intrigues and rumours, in-fighting, power-games, triumphs and dishonorable defeats associated with routine academic peer reviews. The idea that outside society wants an evaluation of their work based on an independent, external audit, using criteria that are different from their own, is antithetical to Mode 1.9 Thus, whereas Mode 2 is characterised by external accountability, Mode 1 is typically introvert. As a source of political struggle and tension, the factor which distinguishes Mode 1 most from Mode 2 relates to how agendas are set – and who decides on them. Thus the parole of Mode 1 is academic freedom, i.e. the sovereignty of academia itself to decide on its agenda. In this, the metaphor republic of science is invoked, implying that academic peers, as citizens in this republic, are eligible to participate.10 The ideal is that questions posed by the research should without prejudice govern what is put on the agenda – only those close to the unfettered research activities are eligible to have a voice in the agenda-setting process. In contrast, in Mode 2, users and interests define the agendas. This, of course, in modern, complex and knowledgedriven societies, covers a broad range, thus the heterogeneity of Mode 2 also reflects the heterogeneity of those who decide on its agendas. Thus, in summing up, Michael Gibbons' idea in The New Production of Knowledge may be synthesised as claiming the existence of two qualitatively different, yet equally significant domains, or 'worlds' representing distinct knowledge production systems: !

!

Mode 1, which may also be termed an explanation-oriented world, which is based on the universities and academic culture. Organised in formalised, disciplinary and authority-oriented entities, the strength of this research is building solid knowledge based on explaining 'why'. Their bias is a propensity to produce knowledge based on ex-post explanations. Mode 2, which may also be termed a solutions-oriented world, in which the research-agenda is 'how&now'-oriented, primarily in exploiting emerging technological or market opportunities created by scientific break-through, typically undertaken in the

Doing innovative ICT-research

R&D-labs of ICT-industry. Their main objective is to create solutions, making things work successfully, but in the process, they also create new knowledge.

Rethinking the agora? In the community of research on science and technology, there was some excitement one year ago, early in 2001, as the news spread that Gibbons and his team had published a follow-up book (this time with Helga Nowotny as first author) with the poignant title, Re-thinking Science: knowledge and the public in an age of uncertainty, (Nowotny, Scott & Gibbons 2001). One reason for this was perhaps expectations that this book would provide the world with harder, more stringent analyses and more empirical material to support the ideas of Mode 1 and Mode 2, thus making the message from the first book more comprehensive and solid. However, these expectations were not fulfilled. Although they maintained and defended the original conceptual framework from the first book, i.e. the dichotomy of Mode 1 and Mode 2, and the significance of context, the new book introduces a new dimension: the interdependence and symbiosis of the two modes, taking an apologetic position praising the virtues of Mode 1. Amplifying the essayistic style from the first book, Rethinking Science makes many suggestions as to how Mode 1 and Mode 2, as two disparate knowledge production systems, should 'co-evolve' for mutual benefit. One of the suggestions is the idea of constructing the agora (the Greek word for a public assembly area or a market place), i.e. a mechanism by which Mode 1 and Mode 2 may connect and reflect on their roles and activities, and how they may shorten the gap that now exist between the two modes. In analysing this, there is reason to ask why the book is not more specific, why the vagueness and sketchy uncommittedness of a mechanism such as 'agora'. Still, one may interpret the idea of agora as a perceived need for a synthesis, something novel and different from a compromise between Mode 1 and Mode 2. Turning now to ICT-research, to what was earlier depicted as a unique 'innovation regime' of ICT, one may, using the synthesising formula from above, specify a new category. Instead, or in addition to an agora, what is needed may be termed Mode 3 :

25

26

Researching ICTs in Context

!

Mode 3, which may also be termed a futures-oriented world, is one in which knowledge is mobilised and created for policy purposes, for creating political agendas and plans, which will enable novel services, systems and infrastructures, but also for regulations and 'rules of the game'. Traditionally, the international standards setting bodies have played an important role in terms of creating anticipatory standards, i.e. standards for technologies and services in the future. In spite of their technological outward image and rhetoric, these were truly multidisciplinary. However, these institutions were eroded during the 1980s. Still, there are signs that they will re-emerge. ICT research should strive to become an active participant in this.

Mode 3 as a future-oriented methodology There are two basic assumptions motivating the construction of a new, future-oriented knowledge production mode, or Mode 3 for short. First of all, the possibility of Mode 3 is not utopian; it has a prior existence. In fact, Mode 3 is almost a contemporary phenomenon as it was operational during the 1980s, with a few extensions into the 1990s. As numerous theorists have pointed out, the creation of radical innovations depend on institutional and organisational conditions that are conductive and favorable to promoting and fostering these (Chesbrough and Teece 1996, Dosi 1988, Godoe 2000). As indicated earlier, recognition of this seems to underlie the initiatives of 3GPP and ETSI in conjunction with the UMTS-development. Others would point to the activities of computer hackers, as evident in the Open Source Movement and their creation of the operating system Linux (cf. Himanen 2001, Raymond 1999), which in many ways typify Mode 3. Thus the realism or possibility of Mode 3 should not be questioned – Mode 3 is recognisably challenging, but not impossible. Secondly, a Mode 3 should exist for the benefit of society, not just the industry, in order to function as a mechanism for proactively creating future solutions in ICT. Even if the interests of the industry often coincide with society, this intersection is not obvious and these interests may at times be contradictory. If the latter occurs, this makes Mode 3 even more important. To use an economic term, a market failure now

Doing innovative ICT-research

seems to prevail in terms of creating radical innovations that may be of interest for society.11 As evident in the problems that a market driven construction of a broadband infrastructure is encountering in many countries, the simplistic belief that the market has an omnipotent creative capability is now gradually becoming weaker. A well-functioning, healthy market economy has many advantages, however, a salient feature of the market is its lack of rhetoric, its inability to articulate more than 'yes' or 'no'. Thus, the market is not very good in articulating or specifying expectations and ambitions of the future; this inability explains why the market for market analyses and trend speculation, even fortune-telling, is so large. In our highly specialised societies, the democratic institutions are for various regrettable reasons, not able to provide the needed attention or guidance in the shaping of a future-oriented world. Ideally, this is a matter for democratic participation and should be governed by democratically elected institutions, ultimately national assemblies. The attempts to create quasi-democratic institutions, such as the so calledconsensus conferences (cf. Søgnen 1997), in which non-experts are chosen to participate as token representatives of the people, lack a fundamental political legitimacy. Critics may rightly claim that the opinions expressed from such fora are easily manipulated by the experts who arrange these events. Furthermore, the questions they consider and their 'rulings' are typically reactive, similar to the voice of the market: they say 'yes' or 'no', they are not proactive. In short, neither the markets nor quasi-democratic institutions seem to be capable of providing inputs to a future-oriented research for the benefit of society. Thus, the alternatives or substitutes that have emerged subsequent to the erosion of the previous innovation regimes are not promising. The two assumptions create grounds for ICT-research to become involved in the research that is future-oriented. If this is accepted, the nature of the questions may change; asking 'how?' instead of 'why?' becomes interesting. In trying to elucidate this, methodology is of course important, i.e. if methodology is broadly defined as the approach and objectives or rationales for undertaking research and creating knowledge.12 If research becomes motivated by visions of creating futures, the methodological questions become important as questions of social and cultural values and political/normative ideological preferences become more pronounced. In this, the eclecticism advocated in the first part of the chapter may be important. However, there is a strategic aspect

27

28

Researching ICTs in Context

that needs to be elucidated first. In order to have an impact, i.e. to create future-oriented results that serve as a direction or guidance for policy making, planning, investments and decisions, there is a need for a mechanism or institution that mediates these in an influential way. This may have been the idea of the agora in Michael Gibbons' newest book, however, they were not very explicit or prescriptive in advocating this. Thus, instead of starting from scratch in building such a mechanism, the strategy should be to participate in arenas, organisations or institutions having the greatest potential for being influential. As one may predict following the dynamics of Mode 1, academics are not active participants in the arenas that construct our future, at least not in the ICT-world, even if this has changed somewhat during the past decade. However, this claim needs to be modified because there are arenas important to policy making, such as the OECD, in which some Mode 1 actors, in particular economists, are active participants and contributors. Thus, they have become fully-fledged members of national delegations and participate extensively in various expert and technical committees. However, in the arenas defined as ICT-technological, the great majority of participants are engineers or people with natural science education. There is nothing suspect in this, however, engineering is basically a future-oriented project; engineers like to create new worlds and they generally enjoy and welcome people who share their interests.

Eclecticism as a methodological challenge in ICT-research If the context is the most decisive factor in knowledge production, as Michael Gibbons and his team claims, then the proposed Mode 3 is distinct from the others because its primary goal will be to produce knowledge that will be used for decisions pertaining to the future. As described earlier, the structural changes that eroded the unique 'innovation regime' of ICT in the late 1980s are still effective; resurrection of exact copies of these regimes are unrealistic. However, using an eclectic approach to analyse the potentials that may contribute to the creation of similar institutions and mechanisms, may be a fertile point of departure. In this, the attempt is to be a little more specific, compared to the suggested idea of an agora as the meeting point between Mode 1 and Mode 2. We may therefore recast these three modes in new configuration along the following lines:

Doing innovative ICT-research

!

The explanations-oriented world, or Mode 1, has numerous assets and advantages. Foremost among these are researchers' skills and abilities to build solid knowledge based on explaining 'why', i.e. their traditional methodology is conductive to constructing complex systems of knowledge that elucidate phenomenons in nature, society and culture. In a future-oriented world, a Mode 3 environment, this type of knowledge is often needed, because in constructing the future, numerous assumptions are made about humans, gender, society, social dynamics and behavioural patterns, i.e. 'what people really want/need/hate/enjoy, etc'. These assumptions in turn are translated to factors that are important in the design of future technologies, and subsequently, for investment of large resources. Of course, in order to become relevant, Mode 1-type of knowledge has to adhere to its own methodological ideals, i.e. abstain from expressing cultural prejudice and being able to talk the language of this new context. In addition, the Mode 1 system has to create incentives that encourage academics to participate in these settings, e.g. one successful initiative or idea should count as much as an article in a prestigious, international, peer-reviewed journal in terms of academic credit.

!

The solutions-oriented world, or Mode 2, also has numerous assets and advantages, of which the most important may be their goaloriented approach, which is often unorthodox and creative in the implementation of these goals. As evident in the prescriptive management literature on innovations (e.g. Cooper 1996), their methodological orientation is strictly on methods, the yardstick being 'whatever works', not so much 'why' or 'how do we really know what we claim to know'. This no-nonsense approach may justifiably be criticised for its mechanistic instrumentalism, which applied directly as assumptions or design criteria to a future-oriented world, may prove misguided. However, the strength of the solutions-oriented world is its relative lack of prejudice towards novel suggestions and ideas. In addition, they are less status-oriented – thus they would welcome ideas and dialogue with researchers who they feel are interested and constructive. In the relationship between Mode 1 and Mode 2, the latter will probably be more re-

29

30

Researching ICTs in Context

ceptive and open-minded, however, the ideal of Mode 2 is to create something that works. Mode 1 has to relate to this and to be able to use its solid knowledge in a way that is conductive to this. !

The future-oriented world, or Mode 3, in the ICT-context has historically had success due to its previous, unique 'innovation regime'. In this world, by means of innocuous technical standardisation, futures were created because decisions were made for constructing non-existing technologies and ICT-systems, which subsequent R&D had to create. As the traditional arenas for Mode 3 eroded, during the recent years new arenas have emerged which may evolve more fully. In the European context, the EU-commision has attempted to play this role as evident in its 'eEurope' programme and various Framework Programme. In this, they are encouraging and, to some extent, funding various large-scale projects aimed at the diffusion of ICT in Europe. However, the EU itself is not capable of being an 'innovation regime', one reason being that it believes that the market should be dynamic in creating innovations. The emerging, new Mode 3 arenas should become an object of interest for Mode 1 actors, even if these seem impenetrable because they cloud themselves in the innocuous rhetoric of technical standardisation. These emerging arenas are populated by people representing organisations and institutions from Mode 2 – thus their agendas dominate. The challenge of ICT-research, as this has become part of Mode 1, is to reorient itself towards Mode 3, towards becoming an active participant in the multidisciplinary mission of creating the future. This is the main methodological challenge – and the solution may be, at least initially, to adopt an eclectic methodology based on leveraging the best of three worlds.

References Chesbrough, H. W. & Teece, D. J. 1996. 'When is virtual virtuous? – Organising for innovations'. Harvard Business Review, January-February. 65-73. Cooper, R. G. 1996. 'New Products: what separates the winners from the losers'. In Rosenau, M. D. et al. (eds) The PDMA Handbook of New Product Development. New York: John Wiley & Sons. 3-18.

Doing innovative ICT-research

Dosi, G. 1988. 'Sources, procedure and microeconomic effects of innovation.' Journal of Economic Literature, 26. 1120-1171. Gibbons, M., Limoges, C., Nowotny, H., Schwartzman, S., Scott, P. & Trow, M. 1994. The New Production of Knowledge: the dynamics of science and research in contemporary societies. London: Sage. Gibbons, M. 1994. 'Transfer sciences: management of distributed knowledge production'. Empirica, 21. 259-270. Godø, H. 2000. 'Innovation regimes, R&D and radical innovations in telecommunications'. Research Policy, 29. 1033-1046. Himanen, P. 2001. The hacker ethic and the spirit of the information age. New York: Random House. Kline, S. & Rosenberg, N.1986. 'An overview of innovation'. In Landau, R & Rosenberg, N. (eds) The positive sum strategy. Washington: National Academy Press. 275-305. Nowotny, H, Scott, P. & Gibbons, M. 2001. Re-thinking Science: knowledge and the public in an age of uncertainty. Cambridge: Polity Press. OECD. 1992. Technology and the Economy: the key relationships. Paris: OECD. TEP – The Technology/Economy Programme. Raymond, E.S. 1999. 'A brief history of hackerdom'. In Di Bona, C. (ed.) Open Sources – Voices from the Open Source Revolution. Sebastopol, CA: O'Reilly & Associates. 207-220. Simon, H. A. 1981. The Sciences of the Artificial. Cambridge: The MIT Press. Søgnen, R. 1997. Verneverdig eksperiment? – Evaluering av Lekfolkskonferansen om genmodifisert mat. Oslo: NIFU. NIFU-rapport 5/97.

Notes 1

Merriam Webster’s Collegiate Dictionary, 10th edition, 1993, p. 365. These types of rules are particularly strong in cultural notions of food and fashion, i.e. what kind of dishes or what kind of garments that may be combined or belong together. Thus some fashion analysts claim that the fad of combining bulky army boots with very feminine dresses, which was popular a few years ago with young women, had a hidden agenda: This kept older women out of the domain of young women’s fashion because they abhorred this combination, by this the young women were able to create a boundary to older women and their persistent copying of young women’s fashion – at least for a while. 3 For this reason a biography has characterised Reagan and Thatcher as 'ideological soul mates' – Thatcher was credited as being responsible for dismantling the British Welfare State. 2

31

32

Researching ICTs in Context 4

Cf. http://www.3gpp.org for further information. ETSI is an acronym for European Telecommunications Standardisation Institute, located in Sophia Antipolis, close to Nice in France. 6 Prior to the 1990s, the academic community of universities considered ICT as a technical subject, a specialised type of applied research, far away from 'real' science, ICT was not Salonfähig. Being technical, it was considered trivial, commercial, with suspect relations to the military, as evident in the Internet development, where DARPA and R&D-institutes closely related to the militaryindustrial complex played an important role. The shift in attitudes came with the influx of personal computers and Internet in the universities – with this hands-on experience of ICT, academics began to see and appreciate the potentials of ICT. 7 A brief version presenting the essential points of this book may be found in Gibbons (1994). 8 Some analysts have designated this development as 'academic drift' (analogous to 'genetic drift'), i.e. they interpret this as a 'me too'-mechanism, claiming that the rest of the world wants to emulate universities, because their really want to clone universities. However, this is an egocentric explanation because it ignores that most modern societies are being transformed into knowledgedriven economies and modes of organisation. 9 In my line of work, from time to time I am contracted by research funding agencies to undertake evaluations of research organisations or programs, some of these being typical Mode 1. In my contact with these, I am often confronted with the skeptical question: how am I, an outsider to the discipline, in any way competent and capable of making an evaluation? In their world, a fair evaluation would require a person with credentials equal to numerous doctorals in their discipline in order to understand and adequately evaluate them. 10 Needless to say, the ideals are not truly democratic; some of this republic’s citizens are more equal than others, in reality the most equal are full time professors with much power. 11 The notion of market failures justifies funding of public schools, hospitals, roads and high risk research, i.e. funding goods that 'everyone' has a benefit from because these contribute to increased welfare. However, because of uncertainty of appropriating a profit in the future, private investors avoid these, in spite of their desirability. 12 Methodology may be defined as theories of, and reflections on, knowledge. Following this meaning, methodology concerns the procedures and rationales for creating knowledge. For this reason, methodology also deals with how explanations and theories are constructed. Within the natural sciences, methodology is also concerned with how observations in nature are made, and how to construct hypotheses that may be falsified. 5

3 Public places – public activities? Methodological approaches and ethical dilemmas in research on computermediated communication contexts Janne C.H. Bromseth

Introduction: civilisation processes of and on the Internet as context for ethical standpoints The Internet as research arena for human interaction and meaning production is – similar to the medium and its activities themselves – still a pioneer field. As Christina Allen (1996, 177) suggests, 'The cyberspace experiences that can contribute to ethical wisdom are just now developing.' However, as more and more knowledge is produced by an ever growing group of researchers from diverse disciplines, it has become visible that the groups of people using the Internet as well as the medium itself has gone through big changes. Following these developments, the hegemonic discourse about the Net as technological phenomena has also changed. The user group, which previously was heavily dominated by male, university educated, white, middle class, young north Americans is becoming more heterogeneous in relation to social categories like nationality, gender, education, ethnicity and age. Increasing diversity is further existent when it comes to what we use the Net for, and how often. Slowly the societal importance of the medium has grown and become intervowen with existing institutional infrastructure and organisations (Haythornthwaite 2001). Parallell to this, the Internet — as a 'virtual reality', an unfamiliar separate social sphere on the side of the established society, as described by politicians, the media as well as individual users 33

34

Researching ICTs in Context

— is now gradually starting to lose some of its mysterious charisma as it is increasingly integrated into society and more commonly used. Starting out as a 'lawless Wild West', the Net is gradually undergoing 'civilisation' processes, on many levels, by several groups of actors, including legislation (Mann and Stewart 2000). This also goes for using the net for research purposes. After a good decade with humanistic and social scientific studies on net-mediated contexts, the first phase of 'trying and failing' has established valuable knowledge not only in relation to the medium and its users, but parallell to this, on 'doing internet research' (the title of Steve Jones' book from 1999), methodological issues related to the research processes themselves. There has certainly been a movement from 'anything goes' to developing and negotiating an understanding of 'best practices', based on the experience from the growing number of studies carried out. This is further being institutionalised as we speak, where specific rules for the handling of different kinds of Internet-generated material is increasingly developed at ethical review boards and data protection institutions at local and national levels, as well as within research associations.1 It is within this frame that I focus on the use of the Internet for conducting research on human behaviour, and how the experiences of users and researchers have generated discussions on ethical implications of diverse methodological approaches in doing Internet research. Moving into unexplored land and unknown landscapes for studying human interaction, researchers have been forced to rethink basic issues related to practical and ethical sides of collecting material, to be able develop and apply approaches that work for ourselves and our research goals and which may be ethically defendable in relation to our informants. A particular challenge in these processes has been to define and establish a definition of the Internet as a context for the activities being studied, which is in particular rooted in the 'nature of the medium'. Compared to other mediated and non-mediated communication contexts, the Internet gives the possibility of large scale group communication through written text, with a low level of expression through the lack of editing instances, and where the interaction can be stored electronically (Aarseth 1994). Communication is further often publically accessible for all Internet users, allowing non-participating observers to listen in on a chat group or read a web-stored message archive for a discussion group. In that sense, Internetmediated contexts have features similar to both traditional public discourse like newspapers, being broadcasted to a large audience, and at the same

Public places – public activities?

time they consist of many activities and issues that traditionally have belonged to more private spheres and means of communication. This hybridity has created new possibilities for groups and individuals to meet and communicate across geographical space relatively easily. For humanists and social scientists interested in studying communication processes or other social phenomena, a whole new field of study has developed, based on this new rich source of highly accessible material. The use of net-mediated interaction as research material has, however, raised important questions when it comes to how to methodologically approach this kind of material in ethically acceptable ways. In doing this, it has become apparent that the semi-public mediated contexts of the Internet challenge existing ethical guidelines in the sense that there are no easy answers as to what type of material they would seem to represent as compared to other mediated and non-mediated contexts. How we choose to define a certain type of activity as public or private in doing research on naturally occurring discourse is again crucial for central ethical issues such as the necessity of gaining informed consent from informants to be part of the research, and to what extent anonymising is necessary. Ethics and methodology are in this sense tightly interwoven with and have consequences for each other: a specific ethical standpoint will lay premises for methodological approaches, and choosing a certain way of finding answers to a research question will always imply an ethical perspective. In this chapter, I discuss some of the central general ethical issues in doing online research, and how net-mediated contexts often fall in between the traditional division of public and private, making it necessary to find new definitions for what should be considered as public and private contexts on the Net. I also address some key corresponding methodological issues related to defining and accessing different kinds of net-generated material, and I overview some of the dilemmas raised by choosing one or the other solution to them.

General ethical guidelines: pointing in which directions? 'Ethics', as opposed to morals, refers to 'persons-in-culture. In relation to research, this points towards the obligation of the researcher to a group (her informants as well as the society) (Stanage 1995, in Thomas 1996: 108). In all countries and cultures, researchers are accountable to

35

36

Researching ICTs in Context

a set of principles and laws of how to carry out projects in an ethically responsible way, whether formulated as a set of detailed rules, a set of general principles or both. In a Norwegian context, general guidelines for research ethics in the social sciences and the humanities are published and revised by the National Comittee for Research Ethics in the Social Sciences and the Humanities (NESH). These guidelines are 'intended to help researchers and the research community to think about their ethical view and attitudes, become aware of conflicting norms, and become better able to reach well-founded decisions' (NESH 2001: 3).2 What do our general principles suggest in relation to research in the semi-public spheres of the Internet? Are they useful and appliable in their existing form, and in what directions do they point to central issues of the protection of persons involved in research on computermediated communication? Are new guidelines required on the basis of networked contexts? When taking a closer look at section B of the NESH guidelines, focusing on the protection of persons, we see that it starts out by stating that 'the point of departure for research must be fundamental respect for human dignity' (NESH, 2001: 7, pt 5). This is further formulated into three main principles that form the basis for the specified points in this section, aiming in particular at research in which individuals (those being researched) can be identified. The research process must meet certain requirements to: ensure freedom and self-determination (paragraphs 6, 8, 9, 11, 12, 13, 14 and 19) prevent harm and unreasonable strain (paragraphs 7, 9, 10,11, 12, 15, 17 and 18), & safeguard private life and close relations (paragraphs 12, 13 and 14)' (ibid.: 8). A central issue related to all of these principles and one of the main issues in protecting persons carrying out research, is the necessity of obtaining informed consent or informing about the research, and what this principle actually implies of responsibility for the researcher in relation to informants. In practice this means that before being involved in research projects persons should be asked and freely agree to participate in the study, on the basis of fully understood information about its purpose as well as how it will be carried out, and implied consequences for

Public places – public activities?

the individual taking part. There are however some exceptions to this principle. According to the guidelines, 'observation of individuals and groups in public contexts – for instance in their capacities as holders of public posts – rarely requires informed consent, unless the research runs counter to the need of the individual for freedom and right to selfdetermination.' (ibid., 10) The document underlines that 'in the case of active participation, consent is normally required even when the research involves no risk.' (ibid.: 11, pt 8). Of particular interest in discussing ethical perspectives of doing research in net-mediated context is point 9, about the obligation to inform research subjects about the project, and what type of approaches and methods that can be considered as exceptions to the rule: Observation in public spaces, in streets and squares, can normally be carried out without informing those conserned. However, the registration of behaviour using technical aids (camera, video, tape recorders etc) implies that the observed material can be stored, and thus possibly form the basis of a personal register. For purposes of such registration people must, as a rule, be informed that recordings are being made. (ibid.: 11). This quote points directly towards a much debated core issue in doing, research - particularly 'passive' research - in computermediated contexts, rooted in the definition of the context itself. Is accessible information on the Internet per se a public space, allowing researchers to freely collect material for research purposes without notifying the participants? And what would be considered registration and storing of information in using participants' self produced written words? As to this question, our ethical guidance provides no clear answer. From 1999, however, a separate paragraph has been added to point 12 (The obligation to respect individuals' privacy and close relations) acknowledging that the distinction between the private and the public domain may be difficult to draw with regard to information concerning behaviour which is imparted and stored electronically, for instance on the Internet. Of particular importance is the advice given in using material drawn from such interaction, that 'researchers must give necessary consideration to the fact that people's perceptions of what is private and what is public communication may vary.' (ibid.: 13, my emphasis). In my opinion this is a critical addition, one which I will discuss and return to throughout the chapter for it emphasises paying attention to the users' own experience of the context with they interact within.

37

38

Researching ICTs in Context

Public or private? Moving beyond spatial metaphors For the Internet, as with the introduction of other new electronic media, metaphors are used describe an initial phase prior to our gathering sufficient experience to define the medium on its on its own terms. Waskul and Douglas (1996) discusses the role of electronic media (mass media as well as communication tools as the telephone) in relation to how they have contributed in blurring the traditional boundaries between what is considered public and private, time and place. It is hardly the private home and the public town square (that the Internet space has been compared to by many researchers3 as physical places that divide what should be considered private and public anymore, but as much the social situation we are in. In that sense, talking about the Internet as a 'public space' is highly metaphorical, since what is actually creating a 'space' on the Internet through text, graphics, photos, film and sound is the social activity taking place. Physically, we might be in our office, in front of the same computer screen, allowing us to perform and take part in a range of social activities with different numbers of participants involved, to whom we have different relations and who fulfill different functions in our lives. We amuse friends with jokes in private e-mails, flirt with strangers on a chat, or engage in political discussions on an e-mailing list. As Waskul and Douglas argue, the dichotomy of what should be labeled public and private 'domains' on the Internet, is an oversimplification in the sense that it refers to accessibility, not the experience of participants. Even in physical places defined as part of the public room, there is a difference between the intended publicity of activities taking place within it, and this is also problematised in the NESH guidelines as important to bear in mind. Defining a space from the 'outside', based on access, and from 'the inside', based on participants' experience of the social activity taking place are therefore two different positions that do not necessarily correspond. When users participate in publically accessible activities on the Internet, the physical representation of observers is invisible, and the lack of visual and aural cues seems to lower the awareness towards a lurking audience (Sixsmith & Murray 2001). An important aspect of how to define a certain context in addition to its accessibility is thereby related to the purpose of the interacting group: are they seeking public visibility as a primary goal, or is it rather a consequence of the activity taking place (Eysenbach & Till 2001)? In creating a support group for victims of sexual violence, for

Public places – public activities?

instance, a certain degree of publicity is actually necessary to attract and recruit potential participants to the list. Whether by doing this they gladly agree to allow their communication to be used, without consent, by reseachers because it is accessible is, however, a different matter. Most net-mediated group interactions are actually easily accessible, even if they are not directly available on the web (Ess et al. 2001). Signing up for a membership on a MUD-community, or a discussion list with restricted access, is often just a formality and requires no tests or interviews as for participants' intentions for joining the group. But it signals a sense of privacy to the outside world, drawing a boundary around the group of people that is considered belonging to the group as separate from the great mass of net-users. And it is exactly signals of this kind that Internet researchers need to pay close attention to and to interpret. This requires a high level of context sensitivity in researching internet-mediated interaction, encouraging a participant perspective as a point of departure in choosing research approaches rather than spatial metaphors in defining Internet communication activities. Both according to my own experiences in researching electronic discussion contexts as well the experience of a wide variety of research projects carried out in recent years, learning about what kind of social activities that actually takes place in the mediated contexts is one of the most valuable lessons learned. Our research processes have given valuable knowledge and insight into 'the nature' of a number of interacting groups, teaching at least me that what I was dealing with was not 'like a public newspaper discussion', but a dynamic, living group of interacting human beings. Simultaneously, a number of 'worst practices' of research carried out, with high risk of potentially harming individuals and group, has contributed to increased attention towards participants' experiences of participating in computer-mediated communication contexts (see for instance discussions in King 1996, Eysenbach & Till 2001). A much cited research report in this sense, is Finn and Lavitt (1994, cited in Eysenbach and Till 2001), that used posted messages on a web-based support group for survivors of sexual abuse, without asking the group or individuals and publishing both the groups' original names, as well as dates and times for the messages sent. The reason given for not anonymising the group's name and other traceable information was that 'messages posted on a BBS (bulletin board system) are public information'. Even if participants' names were changed by the authors, attention

39

40

Researching ICTs in Context

was drawn to a specific named group and its location, which violates a primary function of offering a supportive environment to people in a vulnerable position. As King (1996) points out, naming the group directly was unnecessary in relation to the study itself and added potential strain to the individuals participating in the group.

Two dimensions: group access and perceived privacy In trying to combine the two dimensions of accessibility and participants' experiences Storm King (1996) suggests to separate between what he calls 'Group Accessibility' and 'Perceived Privacy' in choosing ethically sound research approaches in studying online groups and communities. The former refers to 'the degree with which the existence of and access to a particular internet forum or community is publicly available information', while the latter 'represents the degree to which group members perceive their messages to be private to that group' (ibid.: 126). As there is no a one-to-one relationship between how accessible a group is and its social purpose, both these dimensions need to be considered. For instance, the usenet discussion network is a highly accessible and public domain, whereas the other end of the scale may be represented by a closed e-mail-based group with no publically available subscription address. In between, MUDs and MOOs,4 are good examples of domains where one often finds both open and restricted access areas. Concerning the level of perceived privacy, an academic open discussion group can serve as an example of a kind of context where the purpose of participating often is to make one's ideas known to as many people as possible. On the other end, we find groups meant to serve as a social support network for people sharing a difficult or marginalized situation. The Internet has played a unique role in creating meeting places for groups of people with serious diseases, victims of violence or abuse, as well as sexual minorities. In particular, it is researchers studying health related phenomena using material from groups with socially sensitive topics who, on the basis of their own experiences, have visualised the ethical dilemmas in relation to these (see e.g., Sharf 1999, Sixsmith and Murray 2001, Eysenbach and Till 2001). Due to the sensitivity of the information shared, and according to traditional ethical norms, they are labelled 'handle with care', no matter how accessible the information is

Public places – public activities?

to the researcher. According to Eysenbach and Till, users' responses to researchers sending out surveys online are often negative, as what they seek is a safe environment in which to share what is often very personal information. Making a group the focus for research can feel threatening to this safety, and it can lead to participants withdrawing from the group or stopping their active contribution. Potentially, this may ruin a supportive network, even if this was not what was intended (see e.g. Reid 1996). To avoid adding extra strain to such groups then demands careful considerations as to how to approach the group (and its individual members) in the research process.

Defining the context: conflicting issues of informed consent and privacy in participant observation Who is the intended audience of an electronic communication - and does it include you as a researcher? (Ferri 1999, cited in Mann & Stewart 2000: 46). As previously mentioned, ethics and methodology are closely tied to one other. Defining the context as more or less public or private, combined with research goals and preferred methods, are crucial issues for selecting a research approach. In using methods that involve active participation from informants, like interviewing and doing surveys, it should be quite clear that informed consent is required from participants, even if it is collected via e-mail or in online spaces.5 But what about observation and registration of naturally occuring discourse? It is not accidental that the study of online interaction contexts has given a 're-birth' of ethnography inspired approaches, studying net-communities as culture (Hine 2000). In many ways this arises because of the possibilities to observe and be a part of a group over a longer period of time, but without having to travel any further than to your nearest networked computer to reach your 'field' of study. In doing material collection related to my own master's thesis, I experienced this accessibility as almost 'too good to be true' (Bromseth 2000). As an applied linguist, obtaining access to group communication processes that were already written, without having to go through all the work with transcribing as in researching oral communication seemed equally amazing. Few published research examples existed in

41

42

Researching ICTs in Context

making methodological decisions, and I determined, after thorough discussions with my advisor, that obtaining consent for observing the groups was not necessary. Since both the discussion groups I studied were a) technically organised as electronic mailing lists, with an additional public message archive on the web, and were b) discussing issues related to traditional public debate (radical politics on one hand and the general practise of medicine on the other), I ended up characterizing them as part of the public domain, 'similar to a newspaper discussion'. Further, it was not my goal to focus on individuals and their stories as such, but on how participants who were situated in a specific context then discursively constructed a discussion culture. I was however interacting directly with the groups and its administrators, as I also used an online distributed questionnaire as well as interviewing the list administrators face-to-face. The survey was distributed to the list members private e-mail addresses after securing consent from the administrators, along with a short presentation of the project and a statement securing full anonymisation of the participants and the groups' names. What I didn't tell them directly though, even if they knew I was conducting research in relation to the groups, was firstly about my observation role as a researcher in subscribing to the lists, and secondly that in addition to questionnaires and interviews, I intended to use direct quotes from postings as part of the material. Even if I had categorized them as public activity, I had a growing feeeling of discomfort for not having asked for consent to use their words as part of my thesis as I got to know the group context better and better each day. At that point, however, after conducting questionnaires and interviews, I felt it was too late to ask for consent as my study was heading towards an end. How awkward would it look to ask for their consent to observe the group when I had already been doing it for over a year, and secondly; what if they said no? The uncomfortable feeling reached a threshold at the time when the thesis was finished and interviews with me about its results were to be published in both a medical journal and a radical newspaper, being sure to reach many of the participants in each group. What would happen now – would I receive furious responses from participants, feeling betrayed and used? As a matter of fact, I didn't, and nor did such responses arise when I discussed my results on one of the lists with the group members after the interviews were published. Looking back, this wasn't just sheer luck, but related to my experienced knowledge about the group con-

Public places – public activities?

texts; both had a charter and an aim of operating as part of the public room, as well as in practise, communicating to and with institutions outside of the groups in different ways. Their publicity was addressed explicitly and discussed as part of the group interaction, on the Doctors' List often as a reminder from the list administrator to the group though, to be careful with distributing patient recognisable information. Additionally, many members of the political list, the list administrator in particular, related strongly to the discourse of the Internet as 'a public place where all information is free and belongs to everyone' (see Bromseth 2001 for illustrations). In sum then, it was the lists' technical organisation, their subjects as well as how they would refer to themselves as public fora that in the end made me able to complete the study as it was, and to feel able to legitimize the approach I had chosen. My point is that most of this knowledge was generated as I observed, making it difficult to draw a line for when my preliminary browsing actually became focused research, and in reaching this point, making it difficult to ask for consent as I felt I had already passed that line. In making my full research intentions explicit to the group members from the start I would have avoided this situation as they would either have accepted or rejected my reasons for being present, and I wouldn't have had to feel like a lurking thief with indecent motives.

Like a public town square? What was it with these specific contexts that made me feel so uncomfortable about not asking the groups for consent to observe and to quote their words, in spite of the public character of these online fora? Looking retrospectively at my own process, as well as relating to other researchers' experiences and points of view, I will list and reflect upon some of the key issues related to doing qualitative research of net-mediated group interaction6. How do the mediated contexts they take place within differ from conducting observation studies in a public town square, and what dilemmas arise? 1) Validity: What does qualitative research of human group interaction in context require to be considered credible and what would this in turn imply for informants' privacy? An important element in doing a qualitatively oriented case study of communication patterns in specific groups like the two I studied means that

43

44

Researching ICTs in Context

the context in which interaction takes place is crucial to be able to interpret the meaning of the utterances/text that is produced. Using a discourse analytical approach as I did, looking at how cultural codes for discussion were created in the groups, would actually be meaningless without contextualizing, as linguistic choices and their functions within a particular socialcultural context are closely related (Crawford 1995). For the interpretations and analysis to be convincing then, it was in my case necessary to include relevant information about the subject of the list, the background for their origin, characteristics of the groups of participants, organisational structure as well as how participants were related to one another outside the discussion list itself. The members on my lists consisted of people from a Norwegian, largely one-national culture, in addition to representing a professional local culture and a small political community, as compared to an 'online community' only, with participants from many cultures and countries whose only meeting point would be on the net. Within a small national culture such as the Norwegian one, there is a clear maximum limit as to how many active discussion lists for each subject that actually can exist parallel to each other, and the number of general practitioning doctors and radicals is also small. So even if I changed the names of the groups as well as of individual participants, the risk for identification of both were high. 2) Anonymising: can quotes be traced back to individuals? Not only because of the small cultural contexts would identification be possible, but additionally, when the messages are actually stored electronically on the web as in these contexts, the information used as basis for interpreting was available even after finishing the study. As the list administrator of the Doctor's List in a private e-mail commented to me after reading the ethical reflections in the thesis I had sent him, agreeing with me that consent should have been obtained amongst others because: '... the anonymisation is illusory. Anyone could identify all the involved persons by using the message archives' (private e-mail, my underlining). As Eysenbach and Till (2001) also point out, in using new search engines, writing in excerpts from a message can be enough to trace down the group from where it came as well as the specific message itself, if the group postings are available on the web. In contexts where participants are required to use their real full names, as was the case with the lists that I studied, a simple search would reveal the author of a quoted message directly.7 This leads me to my next consideration.

Public places – public activities?

3) The use of verbatim quotes, and ownership: whose words are they anyway? The issue of using direct quotes of participants in the research report is complex and with conflicting aspects, many of which is highly related to the researched electronic group context. Who do the written words of participants in a specific group 'belong' to – the individual participant, the group only – or 'everyone', and does everyone include the re-use of those words for research purposes without consent? As Stine Gotved (1999) points out in her ethical considerations related to her study of a usenet group, discussing the literature of Tolkien, she could easily have defended not asking for consent to quote from the group, in that the group was situated in the most public sphere of the net. Gotved's decision to be open about her research purposes was in particular driven by her 'sensing a private atmosphere in the newsgroup that made me doubt the definition of it as a freely accessible and observable room' (ibid.: 66, my translation). Using a theoretical concept from Lofland (1998) Gotved offers a unique description of the context as a local third sphere that lies between the public and the private. She also argues that there is a difference between sending a message to a group of peers, in spite of public accessibility, and the 'psychological situation of being interpreted by persons with other purposes than the group itself' (ibid, 66, my translation). Beyond actual privacy or publicity there is something we can conseptualise as the participants'and groups' perceived privacy (as also pointed out by King 1996). Most research experiences report that the intended audience of a message is not the big mass of unknown readers, but the persons considered to be belonging to the group. In oral group communication, the boundaries and positions in a group are more easily defined. If there are eavesdroppers outside of the group that are not really part of it, but are potentially listening in, the awareness of this is of another character as we can actually see it and adjust to it. We are not usually speaking to them as a primary audience however, but to the ones we consider being part of the group (see Clark 1996). In my study, in spite of an awareness of being part of the public domain, postings were often adressed as private messages to the group community, either in general ('dear collegues on the Doctor's List') or to named persons. The list administrators, of course, seemed to have the highest awareness of their public character. Through their positions

45

46

Researching ICTs in Context

they were technically responsible for running the lists daily and had access to who and how many that actually read the groups' interaction through the message archives. The administrator for the Doctor's List expressed in his interview with me that the members' lack of awareness of the group's publicity was his deepest concern, as exposing patient sensitive information potentially could do great harm if they were recognised by patients. Several incidents on this particular list indicate that although the nature of the list was public and everyone knew it, some members simultaneously perceived the list as private. On several occasions members expressed surprise when people outside the group responded to what they had written. One of the members even lost her job partly as a consequence of discussing work related issues on the list.8 The diverse attitudes of recontextualizing a group's interaction, and individuals' messages, in spite of being publically published through archives, was further demonstrated to me while participating on a feminist discussion list, where one of the male subscribers forwarded a number of messages from this list to a different list I was using for my study, a radical political list. His forward was strongly reprimanded by the authors of the messages as well as the group. The feminist list' administrators posted a message to the radical forum, stating that 'such an approach feels uncomfortable and intrusive to the members of the feminist list' (my translation), arguing that the messages were written to this specific audience and not to the group context of the other list and that consent to re-use them from the authors should have been obtained. The insident raised a discussion on the feminist list about the potential consequences of its publicity, and ended up removing its message archive to avoid similar situations in the future. These examples illustrate that in many, arguably even most, cases, messages posted are intended for a specific audience despite the absence of explicit awareness or discussion of this norm. Consent should therefore be obtained to quote to both allow participants to choose whether to be quoted or not, and further, in order to choose if they would want their privacy protected through anonymizing, as in most cases, or if there is a context where being credited for their words is an issue. The latter points towards people who write with publicity as a primary goal for their messages, as in certain academic contexts, or written art, where the subject of copyright would be an issue (Sixsmith & Murray 2001).

Public places – public activities?

4) Research purpose: is direct quoting necessary, and can consent to quote be skipped? There is always a choice of using direct quotes or not. Some researchers argue that reformulating a postings' content should be the main rule in certain contexts, to avoid the problems of making potentially identifiable information public and thereby protecting participants from being identified (e.g. King 1996). This is however a topic that is highly related to research purpose and academic discipline, as pointed out by Susan Herring (1996). In studying language use, as I do, an analysis of interaction patterns without showing the linguistic analytical basis for the interpretations would destroy the validity of the study, as it would be impossible for others to question the relations between the data source and the researcher's interpretations of them. An interesting question in relating to our general ethical guidelines in this case as I see it, is how the use of direct quotes is to be categorized in terms of 'registering using technical aids'. Peoples' written words, as opposed spoken ones, are already materialised and there is no need for using a tape recorder to collect them and further transcribe the recordings. In that sense, defining a text-based context on the basis of comparing them with other technical aids recording oral and visual behaviour can turn out as a sidetrack in the debate about whether informants should be informed or not when quotes are used directly. More important questions are in my view how utterances are used, for what purpose and if this would be in conflict with informants' right to privacy and self determination. Susan Herring (1996) contextualizes this discussion by comparing it to the introduction of the tape recorder as a technical aid in doing research on communication in the early 1960s, that set off a similar ethical debate in research communities. Arguing from a positivist position, she claims that studies legitimately not obtaining consent to quote include: 1) quantitative-oriented studies with low risk of identifying informants and 2) studies that require the use of exact and unaltered naturally occuring discourse. Herring furthers this criteria by including studies with a critical aim, done within publically accessible contexts which she argues justifies the use of direct quotes as well as names of the group itself. In conseptualising 'publically accessible' Herring distinguishes between the groups that are directly accessible on the web and those who require subscription or passwords to access their discourses. Referring to her own previous studies and of one group in particular, a long-term study

47

48

Researching ICTs in Context

using observation and linguistic analysis in list servs with public archives similar to my own, she reasons that the groups have chosen to be public themselves. If the aim of the research is to criticise certain practises within a group, it should be possible for others to access it to see for themselves if they agree with the interpretations or not. Comparing Herring's stance to my own research experiences, where I am obviously highly critical towards my own choice of not obtaining consent to quote, it might be obvious that I strongly disagree with Herring's arguments as well as the premises and reasoning behind them. Firstly, using a positivist paradigm of research is not in itself a legitimate justification. The positivist tradition might hold that the researcher risks validity by notifying a group that it is being observed. Regardless, the point is not only the validity of the study, but the larger picture of purpose and context of research. Secondly, related to my own research experiences, in researching a context with a relatively stable group of participants over a long period of time, operating with their own identifiable names, and communicating through written stored text, has little in common with the public town square scenario, which seems to be the dominant metaphor for Herring's justifications. The gradually appearing knowledge about participants' feelings of privacy and notions of participating in a dialogic group-communication rather than speaking to the public mass of netusers, should then rather constitute an important part of the ground for ethical considerations. I agree, however, that certain critical studies can be defended without informing participants, but rarely without anonymising their identity. Validity can hardly be a reason not to anonymize, as by the time the study is published, the discourse can have changed completely. Group communication, either mediated or non-mediated, are not static social entities, but dynamic and changing. By the time a study is published, the picture can be quite different than it was at the time of the study – old participants have left, new have been added. Even if lists have archives from years back, I cannot see how the principle of protecting privacy can be overruled by the principle of validity here. It should be more than enough to state validity in using quotes as an interpretive unit. Another type of context quite different from the kinds of asynchronous discussion groups that I have mostly been referring to so far, that poses other types of challenges and questions, is that of synchronous

Public places – public activities?

communication forms like chat rooms, MUDs and MOOs. As Malin Sveningsson (2001) reports from her study of a Swedish chatroom, she chose to not obtain consent because the participants already were disguised using pseudonyms, but most importantly because of her focus on linguistic form. If she was to ask participants for consent as she was observing, the intrusion would have destroyed the natural interaction pattern she intended to study in its entirety. As the pace of the communication is fast, and the group of participants changes continuosly, this would have been all she would have time to do. 5) Risks and potential harm of identifying participants and group: how do participants present themselves and how stable is the group? The last point I will mention as particularly relevant in relation to context and obtaining consent refers to the above-mentioned issue of participants' self presentation, along with the stability of group members over a certain period of time. In relation to these two dimensions, my groups could be seen as representing a combination of these factors that would make them vulnerable in two senses. Firstly, as the participants were obliged to present themselves with their real full names when sending messages, the risk of being directly identified through context information, a specific style or through search engines was high. This would be different in the chat context that Sveningsson describes, where participants can choose to use a nickname different from their real name and are because of that less easy recognisable even if the nicknames were published.9 Secondly, my groups represented relatively stable communities in the sense that even if their population changed during the two year period I studied them, a considerable number of users were present throughout the study. Identifying a community as a social unit, either through not anonymising it in the first place, or by providing recognisable or traceable information about it, can then not only affect the individuals taking part, but the whole group as a community. Potential negative consequences of this are reported in several studies of online communities like discussion groups or synchronously organised social networks as MUDs and MOOs. For vulnerable groups with a supportive function, such as Elizabeth Reid (1996) writes about, where users of a group for survivors of sexual abuse were interviewed as part of her study of MUDcommunities, just the fact that attention is drawn towards the group can be enough to make the environment lose its intended supportive func-

49

50

Researching ICTs in Context

tion. In her study, several of the individual participants asked to be quoted directly and some even uttering a wish for being identified by their full name. She chose not to do this however, something she was relieved about afterwards when she received considerable attention for the study, as well as requests from researchers and others to be put in touch with the group through her. This experience illustrates that in netmediated contexts, there is also a question of 'who knows best' how to make decisions about what to reveal of information, both to protect an individual from doing unintended harm to themselves, and to others. This is discussed in the NESH guidelines, as informants might not be able to see the effect of their actions as the researcher does, and this then comes into question when not only an individual's privacy is at stake, but also other individuals that are socially related to the studied individual (ibid, 30). Also, as research results are even more accessible than before because of the possibilities of publishing online, which is used quite commonly by internet researchers, the risks of identification have increased in general, calling for utterly cautiousness in protecting informants. Defining a context and its ethical implication for researching netmediated group communication on the basis of group access and perceived privacy, as King (1996) suggests, are useful startingpoints. However, I find them too static in the sense that they do not consider research purpose and the issue of quoting, as well as cultural and relational characteristics of the interacting group. It is the importance of this contextual information which I have tried to stress in my reflections above. The work of improving existing ethical research guidelines and making new, in particular interdisciplinary, medium-specific guidelines aimed towards an international research community (as the Association of Internet Researchers are compiling) needs to be 'context sensitive' and flexible. To do so also requires incorporating the variation of researchers' disciplinary backgrounds as well as the cultural situatedness of the studied group (also pointed out by Ess et al. 2001). In this way, bringing all our experience with researching diverse computer-mediated contexts in different ways to the table, and taking them into consideration, is of a great importance. How would the different aspects of defining a context, as mentioned so far, be useful in planning a research design for collecting new material? Using my own process as a point of departure and illustration, I will now move on to look at aspects and dilemmas that arise in plan-

Public places – public activities?

ning a research approach when it comes to wether and how to obtain consent to study groups of a more private character.

Moving from the public to the 'private' – changing research position Finding out about own ethical standpoint during the research process in the previous research I conducted made me highly conscious about ethical implications of methodological choices in researching further discussion contexts. As I wanted to move on to look at discussion groups that had predominantly female participants, I discovered that they were a) often organised as restricted access groups outside the web, as e-mail distribution lists without public archives, and, b) often centred around topics of a more private character. Focusing on the contructions of gender and sexuality, I found two groups in relation to which I wanted to carry out research10. One of the groups targeted to women who want to become pregnant, organised as a web-based, publically accessible group with message archives; and one was a group for lesbian and bisexual women, organised as a restricted access e-mail distribution list with no public archives. How can these specific contexts be defined as to whether or not it is necessary to obtain consent and on what grounds? In addition, what implications would different ethical choices have for my own research position and approaches? Obviously, most professionally trained and practising researchers would agree that the restricted access group for lesbians and bisexuals would require informed consent, both on the basis of access, not being situated in the 'public domain' of the Internet, and also because the partcipants would be considered a vulnerable group of people, belonging to a sexual minority. But what about the web-based group for women who were trying to become pregnant? As I have illustrated throughout the chapter, it would be possible for me to choose and to legitimise several strategies, each rationalized and justified with a different sets of arguments. As I try to categorize what separates these positions, there seem to be two main paths of reasoning, related to who is more responsible for individuals' choices of revealing personal information and putting their privacy at risk: the individual herself, who freely participate in a publically accessible forum or the researcher who intend to use this information for other purposes than it was meant?

51

52

Researching ICTs in Context

The answers to these questions can lead the researcher along distinctly divergent paths. For example, Susan Herring (1996) (and partly Mann and Stewart, 2000), would argue that as this group has chosen to organise themselves in the public part of the internet, consent would not be required to observe and quote their discourse. Herring thereby leaves the main responsibility to the individual and to groups to protect themselves from harm; groups can organise themselves privately should they wish to do so (Herring 1996: 166). On the other hand, other researchers (e.g. Sharf 1999, King 1996, Thomas 1996), as well as the NESH guidelines, lay weight on the researcher’s clear ethical responsibility in the process, to consider the risks involved, based on all information available at the time11. Particularly central to this evaluation, both according to NESH and the researchers of health related issues, is the sensitivity of the topic and the nature of the group discussion. This is specified in the NESH guidelines as 'questions concerning sickness and health, political and religious views and sexual orientation' (2001: 13). Let me review the complicating ethical factors in a seemingly simple example, using the pregnancy group. Having difficulties in becoming pregnant is definitely a health related issue. Even if such difficulties do not necessarily represent a 'stigmatised group in society', it can certainly be experienced as a strainful situation for some either over a shorter period of time, or for others, over a period of many years. In applying the list of contextual considerations I argued for above, my approach would still be qualitative, including recognizable context information about the list, and this list is also one-national in a small country. Full confidentiality can not be secured as I intend to use verbatim quotes, which can be traced through search engines (a factor that in itself would require informed consent following Eysenbach and Tills' considerations (2001)). Contrary to the makeup of lists I have used previously, not all participants are directly identifiable, as all participants use nicknames in the discussions. There are, however, a considerable number of participants that have composed presentations of themselves on the website, or have made their e-mail adresses known, reachable just by clicking their online names. As for the stability of the group as a community, it certainly exhibits features we have come to recognise as characteristic of online communities, with a central core of participants having been present for years, while another part of the group consists of women participating for a shorter period of time and is thus more transient.

Public places – public activities?

Striving to minimize the risk of violating privacy and causing harm, the sum of all these factors about the group and the participants, combined with my research approach, on one hand, would indicate that I should obtain consent to study the group. On the other hand, the group has chosen to be part of the public domain, and could have organized themselves more privately. But for what reasons, and who chose the organisational form? There are obvious advantages of being situated as a web-based group, as it is easy to find, easy to use and provides links to other resources all in one place. Other participants' experiences which might be helpful are stored in the archives for new members to browse. Moreover, organized as a loosely connected temporary social network, the group might not be stable enough to act in a co-ordinated group fashion to change the organisational choices made by the initiator. Why must they choose a domain we researchers define as private, just to help us determine the boundaries to keep us out?

From easy-access Mecca to methodological has hassle? There are at least four obvious main issues in preparing a functional research design that will be affected by choosing to obtain informed consent or not: 1) How will the research purpose be affected in informing group members? 2) What research strategies would be possible? 3) How and can informants' privacy be protected and 4) What effects on the group as a social unit can occur? The choice of not informing the group has some advantages related to the research process itself. Firstly, in avoiding 'the observer's paradox', and thereby not affecting the group communication in its natural environment, and secondly in not disturbing the interacting group and its social purpose. The main ethical consideration in not informing about the ongoing research is primarily related to the potential consequences of harming individuals or groups after the study is done, in publishing the results from it. Deciding that obtaining informed consent is necessary, the implications are to a larger extent related to the process of collecting the material. Firstly, it will naturally raise an awareness and have some sort of effect on the studied discourse in itself. As Sixsmith and Murray (2001) exemplifies, it can for instance result in central participants withdrawing from the group, or even removing what they have previously written from the message archives, which would severly affect the basis for interpreting the material.

53

54

Researching ICTs in Context

Secondly, in using interacting groups as focus for research, as compared to doing interviews with single individuals, there is the potential of doing harm to the social function of the group itself. This is reported by both King (1996) as well as Eysenbach and Till (2001). Referring in particular to health related discussion lists, requests from researchers seem to be increasing in frequency on particular topics, and some groups considered these to be intruding on the group's purpose and were therefore unwanted. How to inform and ask for consent in a certain group context without intruding unnecessarily is therefore an important matter. It is connected to both the purpose of the group and group norms as well as to its technical organisation. Eysenbach and Till (ibid.) suggest two main strategies. The first aims at informing the group in advance of the study, which is described as more intrusive. The second implies that the participants whose postings are intended to be used in the research report are asked for permission retrospectively. In preparing my own study, I have ended up with deciding on asking the groups for consent in both cases. Moving from a safe, non-intruding observer position from the outside, the methodological implications and dilemmas raised as a consequence of my ethical considerations have posed some challenges without any straight forward and simple solutions to them. There are some major differences between the way the groups are organized that have forced me to choose different approaches, related to to what degree their users can be reached by me as a researcher. Firstly, getting in touch with the groups and the individual members, and securing that each member receive the information, is a lot easier when they are organised as an e-mail distribution list, wether this is done through a posting to the list itself, or by sending each member an individual e-mail.12 By sending notifications on a regular basis, it is possible to secure that most of the list members would be informed of my presence, intentions and implied consequences for the participants. It also opens up for a potential dialogue with group members that Sharf (1999) had positive experiences with in her research of a discussion group for women suffering from breast cancer. This can further have a good effect when using additional methods requiring active participation, as Stine Gotved (1999) reports from her study of a newsgroup. She received an unusually high percentage of answers to her online distributed questionnaire. Reflecting on her choice to make herself and her research intentions visible, she credits the amount of answers partly to her interactions with the group on their premises, assuming that this made the project appear serious and credible.13

Public places – public activities?

In informing the web-based group, though, there is no control as to how many of the participants that would receive my notifications of the study as only parts of the group are available per e-mail. To secure a considerable number, I have chosen to send messages to the list on a regular basis, but to avoid intruding too much as the group has a supportive function to a partly vulnerable group, I have additionally chosen to make the project known through the web-site's start-page. This further requires co-operation with the administrator of the site, an instance in the group that can be helpful to cooperate with in preparing the study as she would know the context better and have advice on how to proceed. Secondly, I have decided to ask for consent to use direct quotes from individual participants as the research progresses. This may cause few practical problems in relation to the e-mail distribution list, as I have access to their e-mail adresses as they appear in the headings14. In the pregnancygroup though, a dilemma that arises is again my lack of control when it comes to being able to reach participants through e-mail to obtain consent to quote. As a consequence, I have chosen to solve this by only quoting directly when informed consent can be obtained. Because of their situatedness on the web, quotes can be traced back to the group and thereby reveal the identity of both the participant and the group itself. Agreeing with Eysenbach and Till, this is a fact that participants should be made aware of as a basis for deciding if they would accept to be quoted or not. In deciding on how and why to obtaining consent for research purposes, it is important to acknowledge that using groups as a unit of study poses some challenges as to who should give their consent and what it is that they to should then do. As Herring (1996) points out, getting every individual group member to agree on the presence of a researcher would in most cases be impossible. Asking the list administrator only would not be a satisfactory solution either, as this is only one member of the group who cannot speak for everyone concerning such issues. Eysenbach and Tills' reflections on this matter are interesting, when they suggest that informing the whole group in advance of the study would allow individuals that were opposed to it a chance to withdraw from the group, and then seek informed consent from quoted participants individually. Thinking of the one-national discussion group for lesbians and bisexuals in particular, I have given a lot of thought as to: Whose rights are most important in being present in the group, and who should withdraw: group members feeling uncomfortable or insecure

55

56

Researching ICTs in Context

about my presence, and thereby loosing an important supportive network, or me, as a researcher? There are no easy general answers to these questions. In my own case, I intend to ask the group themselves when it comes to the closed lesbian and bisexual group, as they seem to have a welldeveloped community mature enough to make a collective decision as to whether and how I might be present and what they think would be fair. Obviously, there are several methodological challenges related to the ethical choise of obtaining consent to do research in a net-mediated group, as compared to observing silently from the shadow. But simultaniously, in interacting with them on their premises they might tell me and teach me other things that I wouldn’t be included in if I was just listening, and thereby improve the quality of the research. And hopefully I can teach them something too, by stimulating to meta-reflections on their own participation in a specific mediated group context. These can be useful in further discussions on what these contexts are experienced 'as' by the people taking part in them. Only in 'engaging in creative ethical work', as Christina Allen encourages us to (1996: 177), including users reflecting with us, can we as researchers develop functional and ethically sound research strategies on the diverse mediated contexts of the Internet. It might be more of a 'hassle' and time demanding than choosing not to. But all considered, why should using the Net for researching human behaviour be any easier than other types of research contexts? In the end, if we don't listen to participants' points of view in this matter, it will only have 'a boomerang effect' on internet research as a field, influencing both the credibility of the results of our studies as well as affecting possibilities for future research (as also pointed out by King 1996). What group will open their doors for researchers if our reputation is frayed?

Conclusions In this chapter I have tried to highlight the importance of context sensitivity in defining ethical implications when choosing research approaches in studying computer-mediated group interaction. Using both others' and my own experiences from research on electronic discussion lists, as well as the NESH' ethical guidelines for humanistic and social scientific research as a point of departure, I have in particular wanted to

Public places – public activities?

shed light on questions related to informed consent in the ethical grey zones of doing participant observation of naturally occuring discourse. Due to the diversity of the social activities mediated through the net, as well as how they are organized and who takes part in them, internet research must be context sensitive. Also, because of the interdisciplinarity characterising the field of internet studies, additional ethical guidelines must be sensitive to researchers' different purposes and approaches in using the net as a field of study. Our existing ethical guidelines already provides us with good tools related to the ground principles of protecting persons involved in research: Can the research harm individuals and groups or relations? Is the right to freedom and self determination at stake? This should be the point of departure in further developing medium-specific guidelines, not rigid definitions of 'public and private spaces'. Research ethics is not about avoiding 'being busted' for breaking the rules, as often seen in journalistic approach, but implies an obligation to protect informants (Thomas 1996). A crucial issue to be able to do this in future human research of the social activities of the internet is precicely to learn from the experience and knowledge developed, in order to educate researchers and research institutions, users and the society in how to deal with the greater risks posed by new electronic media (Ess et al. 2001).15 There is a great need for good courses, raising ethical awareness and methodological approaches so that future researchers do not have to repeat our mistakes or re-invent the wheel. Our knowledge must further be passed on to and discussed with the institutions responsible for making decisions and giving advice to researchers in how to approach a project in an ethically sound manner. Most importantly, our ears must be open to users' definitions of participating in net-mediated groups – as theirs should be open to us – to engage in a dialogue rather than a monologue.

References Aarseth, E. 1994. 'Postindustriell kulturindustri'. In Rasmussen T. & Søby, M. (eds) Kulturens Digitale Felt. Oslo: Aventura Forlag. 23 - 48. Allen, C. 1996. 'What's wrong with the "golden rule"? Conundrums of conducting ethical research in cyberspace.' The Information Society, Vol. 12, No. 2. 175-189.

57

58

Researching ICTs in Context

Bromseth, J. 2000. Internett: krigsarena eller Kardemomme By? Konstruksjon av samhandlingsnormer og debattkultur på to norske e-postbaserte diskusjonslister. Hovedoppgave (master's thesis). Institutt for anvendt språkvitenskap: Historisk-filosofisk fakultet, NTNU. [cited 19 March 2002]. At: http://www.hf.ntnu.no/itk/kv_bromseth/hovedfag/index.htm Bromseth, J. 2001. 'The Internet as arena for public debate: constructions and negotiations of interactional norms on two male dominated discussion lists in Norway.' Nordic Journal of Women's Studies, Vol. 2., No. 2. 1-9. Clark, H. 1996. Using Language. Cambridge: Cambridge University Press. Crawford, M. 1995. Talking Difference. On Gender and Language. London: Sage. Elgesem, D. 1996. 'Privacy, respect for persons and risk.' In Ess, C. (ed.) Philosophical perspectives on computer-mediated communication. Albany: State of Albany, University of New York Press. 45–66. Ess, E. et al. 2001. Association of Internet Researchers Ethics Working Comittee: a prelimenary report.Published by Association of Internet Researchers. [cited 19 March 2002]. At: http://www.aoir.org/reports/ethics.html Gotved, S. 1999. Cybersociologi – det samme på en anden måde- Ph.dafhandling (dissertation). Sociologisk Institut: Københavns Universitet. Eysenbach, G. & Till, J. 2001. 'Information in practice. Ethical issues in qualitative research on internet communities.' In BMJ. 1103– 105. [cited 19 March 2002]. At: http://www.bmj.com/cgi/content/full/323/7321/1103 Haythornthwaite, C. 2001. 'The Internet in everyday life. Introduction.' American Behavioural Scientist, Vol. 45, No. 3. 363-380. Herring, S. 1996. 'Linguistic and critical analysis of compter-mediated communication: some ethical and scholarly considerations.' The Information Society, Vol. 12, No. 2. 153–169. Hine, Christine. (2000). Virtual Ethnography. London/Thousand Oaks/New Dehli: Sage. Jones, S. (ed.) 1999. Doing Internet Research. Critical issues and methods for examining the net. London/ Thousand Oaks/ New Dehli: Sage. King, S. 1996. 'Researching internet communities: proposed ethical guidelines for the reporting of results'. The Information Society, Vol. 12, No. 2. 119-129. Mann, C. & Stewart, F. 2000. Internet Communication and Qualitative Research. A Handbook for Researching Online. London/Thousand Oaks/ New Dehli: Sage.

Public places – public activities?

Reid, E. 1996. 'Informed consent in the study of on-line communities: a reflection on the effects of computer-mediated social research'. The Information Society, Vol. 12, No. 2. 169-175. Ridderstrøm, H.. (forthcoming). 'Forskningsetiske problemer i cyberkulturen.' in Tidsskrift for kulturforskning. Sharf, B. 1999. 'Beyond netiquette. The ethics of doing naturalistic discourse research on the Internet. In Jones, S. (ed.) Doing Internet Research. Critical Issues and Methods for Examining the Net. London/ Thousand Oaks /New Dehli: Sage. 243–257. Sixsmith, J. and Murray, C. 2001. 'Ethical sssues in the documentary data analysis of Internet posts and archives.' Qualitative Health Research, No 3. 423-432. Sveningsson, M. 2001. Creating a Sense of Community. Experiences from a Swedish Web Chat. The TEMA Institute, Department of Communication Studies: Linköping. The National Comittee for Research Ethics in the Social Sciences and the Humanities. 2001. Guidelines for research ethics in the social sciences, law and the humanities. Published by NESH. Thomas, J. 1996. 'Introduction: A debate about the ethics of fair practices for collecting social science data in cyberspace'. The Information Society, Vol. 12, No. 2. 107–119. Waskul, D. & Douglas, M. 1996. 'Considering the electronic participant: some polemical observations on the ethics and scholarly considerations.' The Information Society, Vol. 12, No. 2. 129- 41.

Notes 1

The young international research organisation the Association for Internet Researchers (AoIR), for instance, will vote on an ethical guideline for conducting research on the Internet at their yearly meeting in 2002. 2 The guidelines are meant to be 'an aid for researchers themselves' (NESH: 4). Additionally, researchers are bound by specific laws concerning collecting, storing and use of data, and are obliged to obtain permission from the Data Inspectorate as to storing systematically collected personal information. 3 This metaphor is used by, amongst others, Garber et al. (1999; cited in Mann & Stewart 2000). 4 These abbreviations refer to Multi-User Domains and Multi-user domains, Object-Oriented; these are online environments and communities dominated by synchronous forms of communication.

59

60

Researching ICTs in Context 5

However, how to actually obtain informed consent in interviews and surveys conducted online as to the legal requirements in different countries is a different matter. The usual criteria is that consent should be given as a handwritten signature on a paper document. Several internet researchers have obtained permission from their institutional data protection boards to use electronically obtained consent, arguing that the extra work in printing, signing and sending consent-forms would actually lower the percentage of participation to such a degree that it would ruin the study (Mann & Stewart 2000). Mann and Stewart pose several practical suggestions for how to proceed in such a process and outline the advantages and disadvantages of different strategies. 6 Several of these are also discussed by other researchers I refer to throughout this chapter, and represent useful resources as they angle the issue from different research disciplines and positions. 7 Helge Ridderstrøm (forthcoming) reports a similar experience in doing a qualitative study of youth web-pages, where the informants' age conflicted with making traceable information available in their interpretations without having consent from their parents (which again would be difficult if parents didn't approve of or knew about their childrens net-activities…). 8 As she had used herself and her work environment as an example in discussing a general topic related to organising primary health care in a small community, this was interpreted by some of her collegues as 'making private information public'. The incident led to a large support campaign from the list community, as well as discussions in press on consequences of participating in public net-fora. 9 The identification of participants' nicknames and online identities without obtaining consent is however also a topic under much discussion, related to participants' experiences of identity online (Ess et al. 2001). Also, nicknames would still be recognisable by participants in the studied group. 10 The groups havenot not yet been asked to participate in my study, and I would emphasise that the information about the groups as it is shown is based on preliminary browsing over a short period of time, not collecting or storing any material in a systematic way. This can seem to be in conflict with my ethical standpoints as argued in this chapter. I am of the opinion however, that a limited period of preliminary browsing in the search for material should be allowed, and is to a certain degree also necessary to find out 'what is out there'. The problem arises, as I see it, when information is collected systematically and focused, for a specific purpose. As for this project, I am just in between these two phases, and in the work of preparing letters of introduction to the lists I have been faced with the methodological challenges posed by my ethical standpoints, that has required careful thought before moving on to interacting with the lists. It might seem unethical to use them as an example given the background of what I have said previously in the chapter. By the time the

Public places – public activities?

chapter will be published, however, I will have asked the groups, and I have been as careful as possible of not revealing too much information about them here. 11 I emphasise that this is not to say that Herring and Mann and Stewart do not consider ethical responsibility as important. There seems to be a disagreement however, as to what extent the researcher should be responsible for protecting informants when informants clearly put their own privacy at risk by participating in the most public domains of the Net. 12 Most e-mail distribution lists programs make it possible to receive a full list of its subscribers by sending a simple command to the automated postmaster. 13 In this context, however, the negative responses to her 'intrusion' of the group discussions were focussed around the issue of receiving 'spam' – irrelevant, non-topic e-mails. In usenet contexts, the value of not receiving off-topic postings is an important part of the normative interaction culture. 14 In waiting too long to ask for consent though, there is a risk that the adress is not longer in use (Eysenbach & Till 2001). 15 Dag Elgesem (1996) also discusses this issue thoroughly in relation to new electronic media in general, posing a set of general principles about how to minimise violating informants' privacy.

61

62

Researching ICTs in Context

BLANK

4 Locating the Internet: studying ICT is passé in the culture of real virtuality Trond Arne Undheim

Context On one level, this chapter chronicles the experiences of a Teacher's Assistant (hereafter called TA) trying to give a lecture on work, ethnography and Internet. The lecture reflects key findings from my recent Ph.D project (Undheim 2002) investigating visions, practices and experiences with Internet technology among ICT companies. However, there are more complex storylines competing for attention. First, the lecture is is constantly interrupted by its audience. They are physical students in an auditorium as well as virtual participants and interlocutors. Hardly Habermasian, patient rationalists, they argue both with their physical and virtual silences and utterances. Manuel, Sherry, Bruno, Barry and John, among others (all prominent scholars in Real Life, RL), contribute to the profound confusion over the hyperlinking of discussions about cyberspace, online and offline space making and methodological practice. There is a religious aura around the discussion when the participants experience the 'culture of real virtuality'. Thus, the article plays around with interactivity both as hype and as praxis. Luckily, there is no difference between the real and the unreal anymore, one may conclude. However, praxiological semiotics is not that simple. Material objects like laptop computers, pointers, misplaced chairs and confused geographical orders refuse to display their obduracy, but persist in their resistance. None of this is new, but it is quite surprising that chapter

63

64

Researching ICTs in Context

might be contaminated by the same disorderliness! Yet, we know stories about ICT can be controversial and are open for flexible interpretation and that the topic necessarily becomes messy when many voices are allowed to intercept the message. Locating the Internet is not a straightforward thing to do. And it might be too late to try. The topic itself seems to evade clear categorisation. But that depends on the progress of the culture of real virtuality, which in itself is produced by pragmatic choices, as well as it being a contingency of our current stasis in, and preference for, our physical settings and orderings.

Introduction I have gathered you all today, not only as my class of 2002 and as part of the initiation proceedings but also because you are my imagined community (Anderson 1983). It is you I am thinking, writing and talking to, anyway. I'd like to take the opportunity to thank you all for being here. TA regretted the opening statement at once. First of all, most of the class had not even bothered to show up. Second, those who did would most likely direct the most valuable part of their attention somewhere else. He knew it. It had happened before, but never with such a distinguished class. Third, the online crowd would most certainly mess things up. They always did. At the start, there is complete silence. Then TA begins his story. But so far there are no listeners. Not even the remote auditorium in a Swiss village in the Alps called Davos is connected yet. The sparkle of the analogue connection with the adjacent room can be heard as the resonance brings Japan's mighty media mogul Yashiwata to a complete standstill as he has his ear pressed to the receiver on his Docomo driven digital assistant. The machines are restless tonight, Sandy remarks coldly. From her side of the Internet, in Austin, Texas, it is late by now. She is surely there, as much as that is worth for a cyber body. TA always wonders if Sandy is an avatar. She certainly sounds like one sometimes. Soon the machines are again cooperating and more people start streaming into the speaker's auditorium. TA finally has the floor again. We have been accustomed to hearing the Internet is changing the way we think, work, play, learn and love for quite a while now. Companies like Cisco and Telenor, politicians like Al Gore and policy makers

Locating the Internet

like Bangemann (1995) are joined by computer visionaries Bill Gates and Nicholas Negroponte as well as by sociologist Manuel Castells in heralding the new society (Castells 1996). This is a society characterised by the culture of real virtuality, a culture where the virtual has become commonplace and we no longer separate between online and offline experience. Here, TA is interrupted by Manuel, who hurriedly, but with unmistakingly short steps and curiously heavy breathing is heard at a distance, as he is we suppose, arriving in the physical teaching room where TA is present. The sounds are audible only to the people physically present (so far only TA and Richard). The rest of the audience is spared the details, the atmosphere, and the annoyingly sad air quality of the yellow brick wall 'pre-Beatles quality' room without windows. Why bother with being present for a discussion about the virtue of virtuality? Total redundancy is assumed). The 4th floor office in Barrows Hall is only twenty footsteps from the auditorium, and would normally require no effort. The room itself is filled with artefacts, books and other material traces from his 1974 flight from Paris to Berkeley, and also reveals his Spanish heritage by the entire shelf devoted to Spanish partisan politics. But as Russian television flashes their 70s style flashbulbs towards us, we understand what is going on: another interview in realtime, of the most annoying kind. TA is really quite annoyed by now. How dare Manuel be late? After all the articles he has scanned and profiled and all of the hours of sweat, tears and thoughts at the Berkeley library. But Manuel is king. And he walks in at the most unthinkable point. The controversial statement about the seamlessness of ICT and that this would mean it is no big deal. Well, TA thought, what the hell. So, concerning the hype that I was telling you all about. Now, as ICT becomes more seamlessly integrated in our advanced societies one would think that visionary discourse would have become more transparent, or disappeared. After all, what is the big deal? A case in point could have been the small artefacts that co-constitute real virtuality culture yet provide only ephemeral physical traces. There is, in fact, some indication that as the object itself becomes mainstream the fetish around the cell phone is cooling down in some business circles. On the other hand, the phone has also become a distinction marker, an object of fashion, style, class, and culture. So, overall, the frenzy around new ICT based technologies is still there. This is also visible in commercials for Internet subscriptions, software, or computers. If we look to actual

65

66

Researching ICTs in Context

usage, the visionary, or should we say, 'politically correct' views dominate there as well. It is hard to get a realistic picture of what people actually use their ICT for, and what it means to them amidst all the necessary surrounding talk. People want to be seen as forward thinking, tech savvy, professional and responsible. TA took a breather and sat down for a moment. The lecture was well prepared, but it could become a virtual disaster. Nobody present apart from Manuel, everybody present online, yet nothing to hide, really, but also nowhere to hide. This, too, was his final lecture before the initiation proceedings were over. He would be one of them, soon, qualified to play around with realities, licensed word wizard and magician of the imaginary, or as some prefer to think, defender of truth. The excitement had kept him up all night. Trying to think of life in other time zones had not helped for a second. He was dead tired, yet virtually forced to be present. The question then, is where is the Internet headed now. And how on earth are we going to find out? To the researcher, these processes present obvious challenges. Of course, access related issues (access to people and information) are always prominent in research. ICT research presents additional challenges because of the somewhat 'shielded' use of ICT that goes on behind the corporate veils or beyond the political correct answers from parents, children, or adolescents in question (Undheim 2000). Not only serious social scientists, but also politicians would be insulted by that reference. Maybe so, but the slight unease he felt referring to his own work was remedied by the culture of real virtuality – nobody noticed the details as they were immersed in the big picture. As TA tries to turn the page, he can hear some banter from Sherry and Bruno who have arrived late for class. The virtue of the virtual lies in how bytes, modems and things make us discover our own materiality, Bruno whispers and giggles while glancing towards Sherry. What a way these French philosophers have with American women. She herself has not yet finished her MUD session with one of her more sexedup virtual personae and is not in the mood for such French flattery. 'She's a character, really, Manuel thinks to himself as he is looking towards the yellow brick wall contentedly. His concerns are more pressing. No book written this week, so he compresses some time and space by thinking about his 1000-page volume on the Internet to be published by Blackwell in January. In other words, the mood is just right to break

Locating the Internet

the spell. TA decides to try to revive the culture of real virtuality – an idea so far-fetched that it might just be true.

On the reality of Internet experience TA is again in charge. So far nobody has really bothered to interrupt his argument. They have been hyperlinking themselves to other things more relevant to their immediate attention spans. The smell of warm computers fills the room with a cosy 'feel'. The remote listeners are bound to feel left out. What could their own petty computers contribute against the mass of laptop punching fingers gathered in this auditorium? The McLuhan crowd in Toronto, he knew, were all online and eager to participate. But their leading figure, Derrick, wouldn't know how to match this one. Connected Intelligence, Derrick calls it, the mass of distributed brain available on the Internet. Safer, then, to quote Castells who at least is physically present: Reality itself [...] is entirely captured, fully immersed in a virtual image setting, in the world of make believe, in which appearances are not just on the screen through which experience is communicated, but they become the experience. (Castells 1996: 373).

TA continues. Information and Communication Technology (ICT) has a central place in contemporary society. By many accounts - and to scholars, business people, and government policy-makers alike - ICT is a generic transforming tool that has implications for society-at-large. In particular, settings like work and industry, schools, and people's homes are all said to be transformed by ICT. Generally, ICT is thought of as a network-generating and network-maintaining tool that changes timespace relations, that is, makes new relationships between people possible against great physical distances. Given this central place of the Internet, where is it located? TA is especially happy with the last formulation. What a contradiction in terms speaking of a central place on the Internet, or even a location, when the whole point was its pervasiveness. But nobody says anything. The silence speaks for itself. They are occupying their great minds with other more pressing issues, again. What a pity this 'connectivity drive'

67

68

Researching ICTs in Context

is to teaching, TA thinks to himself Teachers and politicians seem to think that a PC connected to the Internet is all schools need these days. Looking out in the auditorium, it becomes clear to him that there is more to teaching than Internet connections. Not even Richard, that unrelentless defender of authentic being (Sennet 1998) seems to be paying attention yet. He just smiles to everyone, Tony, Knut, John, and Michel, greeting him or her as they arrive late for class and enter one by one. Trying to keep it up, Richard is probably insulted since he has not been properly introduced yet. So be it, for the moment. His time will come, a time for authentic attention (Sennet 1998). The first answer is that the Internet is where the people are. And people are social. Thus, the Internet is a social space, if it is any space at all. In fact, the Internet is often viewed as a social enterprise, and cyberspace as a social space. It's no surprise that Internet scholars tend to view the Internet as the most fundamental element of globalization, and often share the view that virtual space is more and more important in society (Turkle 1999; Wellman & Hampton 1999). Indeed, We are living in a paradigm shift, not only in the way we perceive society, but even more in the way in which people and institutions are connected. It is the shift from living in "little boxes" to living in networked societies (Wellman & Hampton 1999: 648). These boxes are for instance neighbourhoods, and it seems evident that 'people usually have more friends outside their neighbourhood than within it', thus are 'networked', and they actually use 'computer networks as social networks'. These networks introduce new social elements, like asynchronous communication, rapid exchanges, complex interactions, own norms, and more extreme communication. Forwarding allows indirect ties to become direct relationships, e-mail is accessible, one-to-many communication is simple, it 'fosters weak ties', sustains specialist communities of interest, and support both purely online communities, as well as those that intertwine computer-mediated and face-to-face communication. What a mouthful! Had this not been a top-notch class he might slow down at this point. But they are multimedial, intelligent and merry – as well as busy. In fact, they embody other interests than their physical presence suggests. They give life to the next point, TA thinks, and continues.

Locating the Internet

Indeed, global networks free workers from local, place-bound constraints: … computer-supported social networks successfully maintain strong, supportive ties with work and community as well as increase the number and diversity of weak ties. They are especially suited to maintaining intermediate strength ties between people who cannot see each other frequently. On-line relationships are based more on shared interests and less on shared social characteristics. […] The combination of high involvement in computer-supported social networks, powerful search engines, and the linking of organizational networks to the Net enables many workers to connect with relevant others elsewhere, wherever they are and whomever they work for. (Wellman et al. 1996: 231).

Barry has started to pay attention now. His quick eyes and noisy laptop have stopped interacting for a while, and even the numbers are laid to rest. He seems to be completely offline right now. Ready for immersion, though. Barry suddenly bursts out laughing, and says, Nobody told me we were going to be in RL (Real Life) today. I would have prepared a speech. Indeed, Howard and I talking about the live version of Rheingold (1993), have decided that virtual communities are the real way to communicate. We are pressing to outlaw all other communication. It is just too time consuming. Don't you all agree? At this point even Sherry looks up from her MUD. This is just a bit too extreme. Barry, you must take some more cyber therapy. You seem to have lost yourself on the screen. That was not what I meant by my book. I never thought people like you would misunderstand. I was talking about 15year olds and identity crisis. Adults should meet, talk and then have dinner, dance like we were 15 and have lots of red wine. Richard chuckles. This is bordering on what I once called the tyrrany of intimacy (Sennet 1976). Could we not talk about something else? TA saves them all by continuing his tale. Paradoxically, computer networks also encourage the formation and strengthening of local ties (Wellman & Hampton 1999:650). The last point is interesting, but has been toned down in Internet research until quite recently. The globalist pretension made scholars overlook the local phenomena inherent in all media; that it strengthens local ties. TA thinks about mentioning some recent work by Miller & Slater (2000) on how the Trinidadians use the Internet but leaves it for now. For

69

70

Researching ICTs in Context

obvious reasons. His own experience of Trinidadians was that they went to a dodgy place called the Pelican Inn to pick up foreigners, spoke broken English, and were a bunch of dreamers who greeted you with 'One love' from Bob Marley and ran Internet cafés together with public laundry. So as to wash the global setting away maybe, or just shrink it until it fits with their own cultural texture.' TA stops his own stream of consciousness, and instead takes a deep, censoring breath before he continues. The adherents of virtuality sometimes closely follow those doing research on the Internet. Typically, they have the notion that most of the world is now virtualised, or turning virtual. Social scientists advocate 'Virtual Society' programs designed to check out the virtual world by means of research, computer scientists claim the world is more and more dependent on computers, and there is a whole new field of e-learning companies who advocate virtual learning. Implicit assumptions in virtuality discussions include that information and knowledge is the same thing, that the virtuality trend necessarily will continue (which is a technological determinist statement), that there are few costs with the digitisation of information, and that knowledge can be 'stocked'.' Are you with me? TA looks up from his notes and looked straight at Manuel. He just shakes his head and says resentfully, this is plain wrong, TA. You can't compare social scientists to business people that way. To this TA retorts, Will you let me finish here? It now appears TA is trying to start a battle. A battle of realities. TA fires away. In the computer-assisted cooperative work (CSCW) tradition, computer scientists, information scientists, communication scholars, psychologists, and anthropologists together with corporate and governmental representatives work on the consequences of Internet on work. Their view on globalisation (as space/place) is contingent. The use of online workgroups, discussion groups, intranets and email lists can meld spatially dispersed coworkers into more densely knit, socially cohesive organizations (Wellman & Hampton, 1999: 652). However, despite the development of cutting edge technologies, the CSCW community acknowledges that systems have not met with a great deal of success (Heath et al. 2000: 304). There is, however, little research that shows The ways in which different forms of collaboration emerge, coalesce, evolve, and fragment". And hence how individuals "in concert with each other use various tools and technologies to assemble temporary forms of

Locating the Internet

co-operation, so as, for example, to develop a particular niche in the market. (Heath et al. 2000: 304).

You are being way too critical of CSCW, says Barry. We never had those ambitions you portray here. This is simply misleading. If these virtual grounds had some proper courts, I would file a restraining order against people like you who ruin our movement. What are you trying to accomplish, young fellow? Do you want to tear us down from our pedestals? Come on, will you, I am prepared! TA has never felt more confident. Their words are battling now; a preInternet face-to-face atmosphere reigns, though it might only last a moment. For obvious reasons, Barry is agitated. His body is trying to get an 'optimal grip of the world' as Merleau-Ponty (1962) remarks, trying to find the perfect focus, where to be at a given moment. It is obvious that the auditorium feels uncomfortable now. Sweat pearls start to show on both TA's and Barry's foreheads. TA raises his voice and almost shouts across the room (as well as into the Internet and straight out on Public Radio). The adherents of Cyberspace join forces around the Internet as a liberatory instrument. Indeed, many of them are neolibertarians (Borsook 2000) who in the Internet found a way to express themselves, found a way to make their special interest into the mainstream discourse about what society was about. The precursors of this thought did come out of computer circles around universities, most of which came to be called 'hacker' communities, some remained in the academia, some went on to business, yet others remained in Cyberspace. Well, remain is not the term, because they actively created what now is known as cyberspace through seemingly small virtual communities that were picked up by novelists like William Gibson. As there are no protests, TA continues. Cyberspace as a social theory subscribes to the notion of disembodied communication. Free from the constraints of bodily discourse, the cyber is allowed to operate in a space of possibilities, seemingly endless in its nodes, networks, and options. But just that seems to be the characteristic of cyberspace, as we know it today. Scattered interest based virtual communities that quickly form, evolve, and disintegrate, almost organically. And constant attempts to structure, restructure, and keep what necessarily can not be kept, but is better left in a free flow of impressions, images, and information.'

71

72

Researching ICTs in Context

Here, Howard, distantly present through his Well account connected to the big, yellow screen, rapidly e-mails the following words that appear in bold letters: SOMEBODY GET THIS TA PROPERLY ONLINE SO WE CAN TAKE A BETTER LOOK. HE'S WORSE THAT WE THOUGHT. DOES HE EVEN KNOW THAT MY FRIENDS FIND THEIR WIVES ONLINE? Now, the image has started to talk back. The screen takes on a force TA has not reckoned with. He shivers, and starts to freeze up. But, there is no choice but to go on with the lecture. Yet to the true cybertarians, there is a common message in cyber discourse. A message sometimes formulated as a message to humanize business communication (for instance the popular website and bestseller The Cluetrain Manifesto and you may see Locke et al. (2000), or as a global movement of computer programmers (the Open Source Movement). Yet, the question of cyberspace and identity is a complex one. In what sense do you actually belong to a web community, or to the possibilities of Internet as a whole? These are the questions of Internet scholars. MIT''s Sherry Turkle is one of them. In her Identity on the Screen we learn that American youngsters find themselves online. The computer becomes a second self that they project both enormous desire, and unwanted qualities onto. To Turkle (1999: 643): A rapidly expanding system of networks, collectively known as the Internet, links millions of people together in new spaces that are changing the way we think, the nature of our sexuality, the form of our communities, our very identities. In cyberspace we are learning to live in virtual worlds. We may find ourselves alone as we navigate virtual oceans, unravel virtual mysteries, and engineer virtual skyscrapers. But increasingly, when we step through the looking glass, other people are there as well. Sherry looks at all of us, that is Manuel, Richard, Bruno and I, and says, What's wrong with being an Internet scholar? And I prefer to be called cyberspace ethnographer, mind you. At this point, TA thinks to himself: 'It's clear that I am critical of such sociographical speculation. Not because they do not have a point, but because they are entrenched in their own territory without looking above and around. Now, on what grounds are all of these wrongheaded? The short answer is that they do not look for the finely grained experiences of Internet culture and rather concentrate on the "consequences",' he decides. The obvious remedy against such a politicised

Locating the Internet

interpretation is to dive into the ongoing practices that support, or are supposed to support this extreme argument. After a moment of thought, he decides to be more diplomatic. Well, I will try to show you, Sherry. The dissertation title What the Net Can't Do (Undheim 2002) rings with anti-program rhetoric, I know. But listen, it essentially looks at two things. First, the way ICT is branded, presented and moulded by the high tech industry and other ICT visionaries. Second, the way ICT visions and products are understood and practiced among knowledge workers. The former was the starting point of the thesis, while the latter emerged during the fieldwork. The dissertation problematic precisely stated is the following: how and under what circumstances can ICT change work-processes in a fundamental way? How can visions like 'work anywhere, anytime', 'knowledge work' and 'nomadic work' be understood? How are they understood by the high tech industry itself?' Well. Let me tell you. These people do not even believe their own visions. I have asked them. Having been this direct, TA stops and looks around him, up on the screen, and into the bytes that are sparkling back at him. So, Manuel, Sherry, Richard and Barry (he deliberately mentions the most face-toface looking people around). What do you think? He has been too brave. Decidedly. Because there is complete silence. Embarrassing silence. He gets serious again with a quick sting from his left cyborg chip, the one that would react to Internetted neuronal vibes. With a theoretical framework built from two main sources: science and technology studies (STS) and contemporary social theory, the key figures in the dissertation are Latour (1999), Callon (1987), KnorrCetina (1999), Sørensen (1998) and Law (1994). The social theorists analysed are Castells (1996), Beck (2000), Urry (2000), Giddens (1991) and Sennet (1998). In addition, more medium-range findings are explored through the writings of Wenger (1998), Saxenian (1994), Wellman et al. (1996) and Hannerz (1992). While this mixed bag of writers bring quite different things to the table, their common contributions somehow knits the Internet, globalisation and work practices together. Here a whole array of things starts to happen. The Internet connection breaks down between Europe and the US, people start fighting in Afghanistan, and a plethora of voices begin to roam through the air. Anthony sends an SMS saying he refuses to be put in this category; Karin calls from Bielefeld to say she feels the discussion is far too

73

74

Researching ICTs in Context

sketchy on the nature of epistemic communities. Annalee says it is time to talk about some real regional phenomena soon, and stop this intellectual banter. Knut e-mails TA to make him get to the point quickly. Bruno wants to say something soon, he tells TA from his heavy gesticulation down in the far end of the room. John sends a fax from Leicester saying his 'gaze' is upon us, finally. Obviously, it has been a mistake to ignore the virtual participants. The machines are agitated now. Or, at least the absent networkers have delegated their anger to these ever-present machines. Theory is futile now, at least this type of theory. It seems too defensive to counter the machines talking back, but TA continues. STS theorists produce the expectation that (a) actors are not determined by technology, rather it sometimes is the other way around (b) technology must be domesticated, that is, made our own and (c) organizing and configuring activities are what constitute settings and stable arrangements like 'societies', 'organizations', and 'epistemic cultures'. This brings about a problem for most contemporary social theory that assumes that 'society' is going in a certain direction, and then tries to account for the forces at play. However, social theory has a strong empirical claim for social changes associated with technology and changing mobility practices. After all, high tech has produced an immense variety of technologies, tools, and practices that play out and influence human practices. The tension between STS and social theory stems from a major disagreement among scholars: whether social science should try to account for stability or change. While the dissertation is sympathetic towards social theory's attempt to theorise the contemporary social setting, it is critical towards the attempt made by extreme sociologists of globalisation. To them, modern society is characterised by a profound change in the relationship between people, things, and places. Strong supporters of this process claim we are witnessing time and space compression (Harvey 1989). Whatever its characteristics, seems clear that mobility, in physical or virtual form, is prevalent. Social theory, however, is divided on the meaning of these changes. Some claim we are facing ubiquitous social space where everything and everyone stand in possible relations (Bauman 1998; Beck 2000; Castells 1996, Giddens 1991; Wellman et al. 1996). Tony sends another SMS to TA's cell phone: Good work. Talk to me after class. Manuel, however, frowns at these generalizations, and

Locating the Internet

shouts, the point is the reverse. The offline places, the people who are constrained to face-to-face meetings only have really ceased to exist as far as I and CEOs worldwide are concerned. I feel that is an issue. TA declines to comment, and tries to defend the place argument. Others argue territoriality takes on a completely different, yet still strong, meaning precisely because of these processes, combined with the strong cultural practices that surround our collectives, in short because of our quintessentially social nature (Sennet 1998; Saxenian 1994; Storper & Walker 1989). But might it not be that mobility visions are factishes, a hybrid of fact and fetish, real precisely because fabricated (Latour 1999)? And, is the relationship between the technical and the social necessarily one of opposition? How can we understand hybrid relationships (material-spatial-technical-social)?

The culture of real virtuality While Castells (1999) could be criticised for his overt fascination for computer culture's generic genealogy and scope, his concepts are clearly constructive. A key notion is that of 'the culture of real virtuality' (Castells 1999: 403). Castells' point is that nowadays, the Internet has entered the tissue of society to the extent that there is no virtual reality to talk about; the virtual has become real in its consequences. So, it is an extreme version of the Thomas theorem: what people believe becomes real in its consequences. The culture around electronic media is communicative, two-ways, multimodal and flexible. Its very networked character constitutes a 'hypertext' of interactivity. In fact: It is not virtual reality, because when our symbolic environment is, by and large, structured in this inclusive, flexible, diversified hypertext, in which we navigate every day, the virtuality of this text is in fact our reality, the symbols from which we live and communicate. (Castells 1999: 403).

There are several interesting things to say about Castells' notion. First, he is using hypertext, Ted Nelson's old notion overtaken by Tim Berners-Lee to construct the World Wide Web in a novel, extended fashion. It now means not only the electronic links that defy traditional hierarchy and order in a text, but it criss-crosses the distinction between electronic and cultural fluids. Secondly, Castells implies that 'we

75

76

Researching ICTs in Context

all navigate' in this hypertext every day. Well, some of us might, but to varying degree, and with different results. In fact, Castells does not discuss the possibly divergent interpretations of this new 'text'. As Grint & Woolgar have taught us, looking at technology as text opens up for flexible interpretations, much like when reading books. So, whereas a book might have its inscribed meaning, its script (Akrich 1992), the interpretations might vary the first time you read from the second glance, and will certainly vary between people who have different background and expectations. The possible factish of 'nomadic knowledge production', true if and when fetichised (Latour 1999), produce the need for case studies from global organizations like Cisco, Telenor, and Telecom Italia. In fact, there might be an interesting interplay of vision making and everyday reality making when new Internet based mobile technologies arrive at the scene. This leads towards a theory of how we are pocketing society, a theory of how society now believed to be portable, and easily dismissed as such, makes globalisation become real in its consequences by ways of the everyday strategic practices of knowledge workers and Internet professionals. Here, Sherry raises her hand. I like this pocketing idea. Where did you get that? Did I write it? If not, I will. TA sighs and remembers why he is the teacher only once a year: the others are really his masters. The fieldwork at the base of TA's study is a set of high tech firms and ICT visionary production materials (advertisements, strategy documents, books) collected in Norway, the US and Italy. In particular, two firms have been closely studied, Telenor and Cisco. The dissertation handles strategies of access, given the fact that getting access to high tech settings is a significant challenge to researchers. It is both difficult to get in and to get out the right information because the elite protects it with gatekeepers and are busy themselves. Methodological issues are dealt with by long-term fieldwork in a high tech setting in California, as well as by more traditional interview methods that are tailored to getting close upon elite practices. Where Latour is sketchy, I want to go deeper. We need to extend Latour's (1999) analysis of the factish phenomena by exploring their mode of production. One way to do this is by way of new concepts like nomadic knowledge work, place making, convincing work, and visionary practice. Please bear with me to make five quick points:

Locating the Internet

First, access is mediated by the invocation of expertise (Giddens 1990). Elites, in a knowledge society perspective (Stehr 1994) become switchers between networks (Castells 1996). Elite interviewing pinpoints the micro dynamics of research work. Status relations, as well as traditional research practices, however, are challenged. To what extent can interview practices be completely virtualised? Second, the tensions between marketing and technological development of mobility enhancing technologies in the Norwegian Telecom Company Telenor are strong. They vary across firms and the settings where visions play themselves out are not easy to find, let alone explain. How are American visions of 'work anywhere, anytime' domesticated? How is the notion of the 'nomad' brought to the forefront? Are engineers seduced by these visions? Third, knowledge workers who operate between community and cyberspace are investigated. The case studies are Awarehouse, Cisco, Telenor, Picostar, Campsix, and Berkeley Incubator - telecom companies, community work-spaces/innovation houses, or business incubators. Analysing knowledge communities, workers, and global practices in Norway and the United States, the aim is to identify the pragmatic space between work practices and technological breakthroughs, questioning the meaning of 'social' and 'physical' aspects of work. Could it be that the nomadic workplace still has a meaning and that computers do not make workers choose either virtual or face-to-face, but allow both? How are pragmatic decisions between those options made? Fourth, innovative regions have knowledge saturation developed through time so that it becomes a cultural, technological, and knowledge-based repertoire. These processes occur in face-to-face relations through time. Work, in this situated sense (Suchman 1987) occurs through and between communities of practice (Wenger 1998). In a sense, the 'setting' constructed in this article is a hybrid of both space and place, but feels like a community still, much like Bourdieu's (1996) habitus explains how practices are embodied habits, structured, yet still active and still structuring. Properly deconstructed as such, then, could high tech practices be considered 'advanced' and 'avant garde'? TA feels alone now. He wishes Lucy were there. Last year was such a treat. His worn-out copy of Suchman (1987) presses against his fingers. Fifth, technological visions are co-produced by visionaries, knowledge workers, and policy-makers - but not only through 'on-the-job'

77

78

Researching ICTs in Context

diffusion. Rather, the 'everyday' of knowledge workers is focalised. Technology as such plays a lead part. Here, I follow Latour (1999) and Fischer (1992). When dominating our 'symbolical environment', the uptake by users and 'abusers' works in such a way that the society of globalisation appears as a readymade and unquestioned reality. The most powerful factish within the factish is 'virtual society' which essentially would give life to Cairncross's (1997) notion of 'death of distance'. But wait, says an incoming British scholar. Talk about coming late for class. He cannot have attended a Public School, TA is sure. But who cares about manners, anyway. Miller & Slater (2000) have shown that Internet on Trinidad is a very local, chauvinist affair with very little Internettish and virtual shebang about it. Look here, for instance, how they argue. And he directs his electric pointer towards the screen covering most of the left wall, allowing a URL to pop up in the upper left corner, and then after a split second, http://ethnonet.gold.ac.uk/chapter.html transfigures into chapter one of their book. Another flick and a ten second film from the 2002 Trini Carnival arouses even the most inhibited of the students present. This is incredibly well timed. TA can feel the excitement as this reference makes the transition to his own contribution swift and smooth. OK, he says. I will talk about that now.

On marginality and the sociology of place making Social reality consists of people, things, and places that relate to each other through specific social imaginaries, or symbolic ensembles, which are constantly subject to place making activities. Place making occurs when people actively participate in the configuration of the places they inhabit and the spaces they touch. My favourite example is the place making of knowledge workers. I maintain that knowledge workers cannot exist outside of a configuration, a sphere of influence that gives value to their work, or legitimises their work as 'knowledge'. I will use Silicon Valley as the example of place making activity involving technology, territory, knowledge, and organisations. Silicon Valley has an intellectual repertoire, a daily life that resembles that of a knowledge worker. This means a knowledge worker is a 'normal' phenomenon in the Valley. His actions are more or less mainstream. There is nothing peculiar about working for a high tech company,

Locating the Internet

owning your own company, or supplying consulting services to the high tech field. In fact, it appears like the most normal thing to do for a young, ambitious person. Now, we do not need a highly developed imagination to claim that it is 'easier' to be a knowledge worker in Silicon Valley than elsewhere. Not easier because there is no competition, rather the contrary; but easy because the institutional arrangements are fit for it, the lifestylechoices are in place. And, the examples of how to do it are readily available, whether in the form of your parents' stories, your friends, your teachers, or just the critical mass of total impulses from the outside 'reality', that of Silicon Valley itself. A visit to Palo Alto, the small, ideal type high tech upper class city heaven with New England porches to have lunch on, and German Porsches to enjoy in the driveway is a perfect way to grasp what I am talking about. Here it is all so visible, laid out before you. In other words, place making often consists of 'doing what seems sensible to do' given where you are. Location plays an enormous role in this choice. It determines what you are influences are, who you meet, what the social imaginary looks like. These fundamental patterns are hard to escape. Place making, thus, is both a passive and an active process. It occurs without notice on an everyday basis. Your 'paramount reality' shapes the type of place making you will undertake. And in some sense, you are made to fit into the place you are at. As Bourdieu's social analysis shows the structure of society, visible in the most highly regarded symbolic capital shapes the actor's choices. But, what this means is certainly not that community is for nicer-than-life behaviour and necessitates a full credence in altruism as opposed to selfinterest. MacIver (1931: 63) argues that: [While the first element of community sentiment is the sense of communion itself] so that when they say 'we' there is no though of distinction and then they say 'ours' there is no thought of division. […] It is rather that the interest of the individual is identified with or merged in the larger interest of the group, so that he feels indissolubly bound up with it, so that in his thought the community is 'bone of his bone and flesh of his flesh'.

Globalisation is not occurring in terms of a shift from place to space, from place to global/local, or from place to glocal (Robertson 1992). Rather, something more radical (in every sense of the word) is happening. We see a move from place to the joint processes of hyperspacing and hyperplacing, the intensification of both place and space.

79

80

Researching ICTs in Context

One way to illustrate the pitfall of one-sided cybertarian views is the work of Internet scholars. While in some ways the virtual encounter may be understood as a marginal situation, that is following Jaspers, the situations where reality is proven through negation; like dreams, deaths, and world-breaking (Berger 1967: 43), we can in no way assume that the virtual takes over the situational definition of most actors. Rather, the marginal is characterized by ecstasy, that is, literally 'stepping aside'. Internet scholars deploy what Knorr-Cetina (1999: 63) would call a liminal approach to knowledge, which is "knowledge about phenomena on the fringe and at the margin of the object of interest". In her account of high energy physicists: [They] incorporate liminal phenomena into research by enlisting the world of disturbances and distortions, imperfections, errors, uncertainties, and limits of research into its project. […] High energy collider physics defines the perturbations of positive knowledge in terms of the limitations of its own apparatus and approach. (Knorr-Cetina 1999: 64).

Thus, virtual community should, properly understood, be a subtext to the community-at-large, that is, our society. This would yield considerable understanding of our day and age. Break. Do you get it? TA is desperate now. The class is frantic. The virtual grapevine passes messages crisscrossing between online and offline class members. He forges on. However, Wellman et al. (1996) and Turkle (1999) end up treating the margin (Internetted youngsters and addicts) as the main praxis. The problem with Wellman et al.'s (1996) and Turkle's (1999) observations is they by now are so immersed in their own study objects that that they start seeing things that were not there before. That itself would be ok if we found they were sources of 'fresh power' (Latour 1999), using the Internet as a laboratory. But most likely they have produced fictional accounts of virtualities that belong to a very limited historical phase of this medium's development, and certainly limited in the story of societies. TA knows this is bordering on insulting both Sherry and Barry, but he takes his chances. The reaction is surprising. They nod silently, but Barry can't resist commenting, so, young fellow, what makes you think you can escape this same logic? To this, TA has no answer. Shouldn't the young Agent Provocateur be exempt from reflexivity? He decides to continue and discuss world making.

Locating the Internet

Turkle (1999) forgets that young people outside of the MIT campus surroundings do other things as well, that Internet is only one constituent of their world making. Also, Internet scholars, it seems, are so defensively arguing the importance of Internetted observations that they forget that the medium has seamlessly introduced itself to most of Western society without us noticing. When we do, rather than find ourselves perplexed, worried, taken aback, we feel comforted, happy, and up-lifted. But maybe not changed in a fundamental way? It is simply a continuation of communication with other means. And those means do have an impact. But the Internet does not operate in a sphere 'outside' society, somewhere ephemeral where we cannot grasp it. And the usual means persist and do not fade away. Just like the radio survives the TV which in turn will survive the Internet. It is indeed very easy to forget that society in itself has always had ephemeral characteristics. No imagination can ever be completely captured. Neither in the tableaux of the French Impressionists; in the machines of Gottlieb Daimler; nor in the words of Shakespeare. Let me remind you: I have set out to study the possible transformations of the 'social' in the wake of globalisation, Internet, mobility, and marketing. I have in no right to dismiss any of these phenomena. But I have every intention to weaken the 'strong programme' of its most extreme adherents. In summary, it seems like a greater challenge to explain why society is stable and apparently unchanged in its fundamentals by all of these imposed conditions rather than to explain the forces behind a change which, at best, we can observe in its generational effects. These effects are only starting to appear, in the case of globalisation, and in the case of the Internet we will perhaps have to wait another 20 years. Durkheim (1973: 138) was already aware of the loosening effects of mobility, yet did not believe it would de-stabilise society altogether: As we advance in the evolutionary scale, the ties which bind the individual to his family, to his native soil, to traditions which the past has given to him, to collective group usages, become loose. More mobile, he changes his environment more easily, leaves his people to go elsewhere to live a more autonomous existence, to a greater extent forms his own ideas and sentiments. Of course, the whole common conscience does not, on this account, pass out of existence.

The grand social theories of cybertarians, or sociologists of globalisation, seem void of detail. Short of specific examples they talk in terms of

81

82

Researching ICTs in Context

'epochal shift', 'flow', 'networks', and 'images' without specifying actors, forces, or social structures that go along with this swift process of change. We are left with musings, examples, anecdote, and not a small portion of apocalyptic worry. At this point, Bruno nods happily. Finally, a sensible comment. Cela me plaît beaucoup. Pocketing society shifts the focus from the technologically and informationally constructed society to the portable weltanshauung created by the combined mimetic/aesthetic, technical, and social uptake of mobility. As a factish, it consists of factlike and fetishlike elements. Interwoven, as it were, with notions of global as not merely immaterial, open, and connected, but also as material, local, limited, and limiting; and the same for local. Together, the local/global, the space/place, and the material/aesthetic are dismantled before a practice based society where individual practice is mobilised in and of collectives. In a way, nothing is new under the sun, for as Durkheim expresses, the individual emerges through the collective. Or, as Knorr-Cetina (1981) says, micro situations have macro foundations. Here, Karen raises her eyebrows (this is seen only to Karen herself given that she has forgotten to turn on the camera on her laptop) and bursts out, this is interesting. But when I think about it only epistemic cultures are able to transform the micro to the macro. Actually, this is what Castells (1996) means also, when he talks about the networkers and the networked. To this, Manuel nods, though as he is more interested in the 'macro' importance of his own books (after all, his trilogy is displayed at a trendy business café in Palo Alto, Silicon Valley). I prefer to spend my time talking to CEOs and Heads of State at the moment, he says. So, maybe your perspective is slanted, TA suggests, but nobody pays any attention to his sidekicks anymore. The audience is listening, TA suddenly remembers how Lucasfilm Ltd. tries to remind everyone that his or her DTS sound system is impeccably real. Here, nobody is listening, but the system is working perfectly. As we then might be able to see, the nomadic stance is also a very territorial one. Like parasites (Zimmer, 2001), nomads are careful with their hosts otherwise they would be homeless. The discourse on virtuality and work is complex. It would be tempting to disqualify the virtual workplace altogether. But certain phenomena persist. Distance work continues to occur; a lot of experiments are going on. In some sense, most knowledge workers are virtual workers. They use some kind of

Locating the Internet

technology to take care of the problem of presence. Whether paper, computers, or groupware, technology interferes in thinking patterns and contributes its own eigenworth to the quality of the working experience. The sociology of place making explores what happens when visionary and other practices by ongoing, precarious attempts configure stable versions of reality. This pocketing process has material, symbolic, and technical components, but can seldom be reduced to any of them. Wait, wait, says Richard. You mean, out of like this class and the way you've tried to establish a common setting for this lecture? Yes, Richard, says TA, that's an example. But like Sennet (1998) is worried about, I think the corrosion of character is quite relevant to what is going on. And nobody talks to another anymore (Locke 1998). The desire is to be everywhere at the same time. So, pocket society, then, refers to, not only the placing and spacing of Internet, workers, and globalisation, but also to the provinciality of this process. Society, as it were, becomes pocketed, brought forth as a portable process, but falls down as a pocket solution to a bag problem. Pocket, also because society is smaller than before, portable, mobile, or what have you. According to sociologists of globalisation, we live in a society that can readily pick itself up and carry on as nomads. Without territory, without fixed points of reference, Pocket Society becomes a fluid society (Bauman, 2000). The 'weird' stance of sociology of globalisation - that society has to be abolished in favour of networks - is caused by a rather sudden discovery of the activity of technological artefacts, information, images, and networks. It is tempting to say that if the theorists only relaxed a bit would discover are things that have 'always' been part of societies, and does not destroy the factish of society. TA knows he should have credited Latour (1999) with this last point, and he looks worriedly in Bruno's direction. Bruno has begun fiddling with some thoughts of his own, and is no longer present. Only his avatar is. Got to go. See ya. Exit Bruno. The machine has traces his footsteps, he's now exactly one block away, yet mentally he's in a completely other territory.

Street opinion of Internet practices If we move away from visionary discourse to consider 'street' opinion on these matters, the result is shocking. TA has arranged its so they are performed by Guru, the same DJ who produced the fusion album Jazzmatazz. The machine crackles a little, and then begins.

83

84

Researching ICTs in Context

ICT is 'everywhere'… Hard to document its uses. Becomes as generic as 'writing', 'reading', or 'meeting informants', activities that most researchers omit from their reports because they are omnipresent. ICT is 'obvious'… We no longer think about it, because it is there. Or, we have stopped thinking about it, and have 'landed' on certain interpretations that we reverberate without actually investigating our own practices. ICT is 'passé'… It has become a topic of the past as more interesting features come along, such as the larger schemes of bioengineering. Partly, these new developments arise in conjunction with a repressive interest for cognitive elements compared with emotional, biological, and spiritual elements that now are entering both research mentality and mainstream society. These tendencies have been there for a while, but arise as we are definitely out of the 1990s spirit of frantic discovery and millennial celebration. ICT is 'disappointing'. It did not deliver what it promised. After all, the visions never seem to come true even though they have existed for 30 years now. We still do not rely completely on computers; we still meet face-to-face when we want to share important thoughts, emotions, or ideas as opposed to organizing some virtual reality session. With the notion that ICT is disappointing also goes the illusion that ICT could bring humanity closer together in one virtual haven. If we now agree that Internet type artefacts have been around for a while, why not try to address that fact? There is, of course, no question that everyday Internet practices are relevant to research. Methodology is a standardised repertoire, a bundle of techniques employed in research, and ICT becomes a part of traditional methodology, influences the collection, study, analysis and presentation of data. In this way, ICT also tends to produce new areas where you can find data. It then adds to the repertoire. For instance, the Internet is a rich source of data, both on individuals, community-based groups, firms, and societies. The whole range of computer-mediated-communication (CMC) occupies a large part of people's daily lives, both professionally and for leisure purposes. In fact, ICT is partly responsible for blurring the distinction between work and play, work and spare time, because it arguably ties these processes together by tightening time-space relations so that you can easily be at work but do leisure chores or vice versa. The social implications of ICT are emerging: they are not stable, static, nor do they in any way following a logical pattern. As we learn more about

Locating the Internet

the way ICT impacts on society, we stop making premature judgements on where things lead. Often, a certain 'technological trajectory' is intercepted by user practices that go beyond what we thought or what the 'technology allows for'. An example is the telephone, which was thought of as a business tool by the original marketer, but in its first decade from 1840-1850 private individuals saw the benefit of calling relatives who lived far away. As ICT becomes part of our everyday life, it becomes harder to separate ICT specific effects from other social phenomena occurring simultaneously. For instance, in 'learning' there are many variables that come into play. Online learning and tutoring has had a mixed reception among students even though it is popular among corporations (who earn money from it) or politicians (who gain points from parents who think their offspring will now catch up with the rest, or gain points in the corporate sector). The relationship between knowledge and ICT is crucial. Most management practices include trying to organise work practices, valueadding information and firm-specific knowledge by ways of databases and intranet communication tools. Despite this, there is a striking lack of attention to how knowledge really flows between people among the leadership of such organisations. Thus, the market for consultants and knowledge go-betweens who can point out new best practices, suggest new ICT tools to be implemented, and remedy organisational shortcomings. The Internet mentality, or with Castells 'the culture of real virtuality' (a slightly stronger point), is an overarching issue and is a somewhat bold thesis, yet highly popular in current social science theories. Here, multimedia is seen as more than a tool in that the tool itself shapes its users. It somehow becomes part of the users much like extended body parts, which by some is labelled a cyborg feature (machine and man at the same time), then enters our thought system and becomes part of what we are as human beings. This strong thesis is upheld by Castells (1996), Himanen (2000), Turkle (1999) and Wellman (1999).

Implications for research practices In closing, I would like to spell out four clear-cut findings from my work. TA is at the end of his lecture, the auditorium smells of sweat, the machines are tired, and the Internet connections are clogged and slow. The old wooden chair becomes a refuge for TA as he prepares the

85

86

Researching ICTs in Context

closing comments. There is euphoria in the air, the magic moment is near. But what will the judges say? The students now have all the power. Their judgement will prevail. TA is finally going to get off stage! !

The Internet is more and more a 'natural' part of how most people in the US and Europe communicates with each other. The consequences of this, however, is that you cannot talk about purely 'digital' solutions to problems like the organization of work, knowledge management and information dissemination. Secondly, ICT is not a field in and of itself anymore, and should not be researched isolated from other types of communication. It certainly does not have its specific, unalterable technological trajectory governed by technologists, multinational companies or politicians alone. On the contrary, all of us contribute to shaping what technology does to us, and we do not really have anybody to blame when the development goes in a certain direction.

!

If we still are to point to a central force in developing, using and talking about ICT we may look at so-called knowledge workers. That is, people who work with information and knowledge in companies – whether they be consultants, technologists, marketers or leaders. These not only 'push' technology through to the market, but also have an equally important function in the more mundane ways they participate in reverberating the technological messages around us. They use the Internet, often buy new technology, talk about technology and bring it into their everyday life. They domesticate technology.

!

The vision that ICT or the Internet will change the mobility patterns of society overall, for instance so that Northern Norway will become equally 'central' as New York City, is an illusion. And it will remain an illusion. This is not because technology does not improve, but because we always seem to want to embody a physical environment as well as we operate on the Internet. It just is that way. When people are involved in place making activities on location, the Internet can very seldom match this. A possible exception is software development, but even there the physical, face-to-face meetings are important.

Locating the Internet

!

The ideas about nomadic forms of work (work anywhere, across limitations in time, place and space) are overly exaggerated. The research backing up such claims is too abstract and too short of empirical grounding to prove the point. If you analyse global, distributed network working in a truly empirical manner, you find that even here the social glue has to be recreated through meetings, phone calls, emails and travelling. This points towards the need to study what mechanisms come into play, how these experiences are shaped, and how the trust becomes recreated through time. Especially important is the notion that development of knowledge is not only a question of dissemination, but also has to do with convincing somebody that what you have is knowledge – true, effective and important. To achieve this we do not only need physical encounters, but powerful presence in those encounters. We need to use all means available, and even then we often do not succeed. Most companies have, in fact, already realized this to some extent, even though their rhetoric does not show this insight.

Conclusion TA has said the big word. A pencil drops and everyone looks at Sherry. To catch their attention again, TA deliberately falls off his wooden chair, and provokes some hesitant laughter by those physically present. The others produce question marks on the big screen, and impatiently await his conclusion. If the Internet has entered society you cannot study it anymore. At least not as 'Internet studies', 'ICT studies' or whatever onedimensional labels we have put on it in the late 1990s. Now is the time to re-evaluate our conceptualisations and move on. My choice has been to largely study the Internet without actually using the Internet. I have looked at how the Internet intersects with knowledge work, and I have tried to look at how visions of how Internet was supposed to work translated into everyday action and mentality among knowledge workers in Silicon Valley, Norway and Italy. The main game, as it turns out, is to succeed with 'convincing work', to create a meaningful assembly of knowledge, experience, things and actions so that people mobilise –

87

88

Researching ICTs in Context

emotionally, epistemologically, or politically. This is done by providing true human encounters, attracting power by minimising threats - creating a setting, and maintaining attention over time. ICT is not a valid topic in and of itself anymore. Just like you cannot study TV without studying specific audiences or programs (Hine 2000), the topic of ICT is pervasive, knit together with culture and technology, and is increasingly in interaction with all other elements of society. Network society (Castells 1996) is about the pervasiveness of hypertext into a 'culture of real virtuality' where there is no 'virtual reality', no separation between them. However, far from saying that the shift is a paradigm shift, I claim this is a pragmatic choice of interpretation, and that it should be, among multimedia researchers. Castells takes the 'culture of real virtuality' to mean that all of us experience the virtual as real phenomena because it is integrated with the rest of our life in novel, complex, and networked ways. I have shown that the same culture could alternatively be seen as a double dynamic. It can both be (a) an intensification of the difference between online and offline space making and (b) a constant transgression of these differences, or boundaries, by the stuff (multimedia, artefacts, people) that surrounds us. Manuel, Bruno (or his avatar), Sherry and Richard clap their hands now, and they all rise. Even the machines are humming contentedly. Occasionally, flares of small smileys move across the yellow screen. Eyes meet. Bodies touch. Glasses are lifted. Fingers cease tapping their keyboards in silent appreciation. Unity at last, may 'the Force' be with us always. The factish seem to have worked. The culture has been enacted. Total collaboration might not have been achieved, but who was to expect that. And from Windows, Bill nods and says he couldn't have done it better himself. A Kodak moment. Real. Virtuality. Now.

References Akrich, M. 1992. 'The de-scription of technical objects'. In Bijker, W. & Law, J. (eds) Shaping Technology, Building Society: Studies in Sociotechnical Change. Cambridge, Mass: The MIT Press. 205-224. Anderson, B. 1983. Imagined Communities. London: Verso. Bangemann, M. 1995. 'Policies for a European information society', in Programme on Information and Communication Technologies (PICT). Charles Read Lecture. Uxbridge: Brunel University. A summary of the speeches. 5-12.

Locating the Internet

Bauman, Z. 1998. Globalisation. Cambridge: Polity Press. Beck, U. 2000. What is Globalisation. Cambridge: Polity Press. Bellah, R. 1973. (ed.) Emile Durkheim: on morality and society. Chicago: University of Chicago Press. Berger, P. 1967. The Sacred Canopy. New York: Doubleday. Borsook, P. 2000. Cyberselfish. New York: PublicAffairs. Bourdieu, P. 1996. Physical Space, Social Space and Habitus. Vilhelm Aubert Memorial Lecture. ISO-rapport, No. 10. Brosveet, J, & Sørensen, K. H. 2000. 'Fishing for fun and profit? National domestication of multimedia: The case of Norway'. Information Society, Vol. 16, No. 4. 263-276. Brown, J. S. & Duguid, P. 2000. The Social Life of Information. Boston: Harvard Business School Press. Cairncross, F. 1997. The Death of Distance. Boston: Harvard Business School Press. Callon, M. 1991. 'Techno-economic networks and irreversibility', in Law, J. (ed.) A Sociology of Monsters. London: Routledge. 132-164 Callon, M. 1987. 'Society in the making. the study of technology as a tool for sociological analysis', in Bijker, W.E., Hughes, T.P. & Pinch, T.J. (eds) The Social Construction of Technological Systems. Cambridge: The MIT Press. 83-103. Castells, M. 1999. 'An introduction to the information age'. In MacKay, H. & O'Sullivan, T. (eds) The Media Reader: Continuity and Transformation. London: Sage. 398-410. Castells, M. 1996. The Rise of the Network Society. London: Blackwell. Durkheim, E. 1973. In Bellah, R. (ed.) Emile Durkheim: on morality and society. Chicago: University of Chicago Press. Fisher, C. 1992. America Calling: A Social History of the Telephone To 1940. Berkeley: University of California Press. Giddens, A. 1991. Modernity and Self-Identity. California: Stanford University Press. Giddens, A. 1990. The Consequences of Modernity. California: Stanford University Press. Hannerz, U. 1992. Cultural Complexity. New York: Columbia University Press. Harvey, D. 1989. The Condition of Postmodernity. Oxford: Basil Blackwell. Heath, C., Knoblauch, H. & Luff, P. 2000. 'Technology and social interaction: the emergence of workplace studies'. British Journal of Sociology, Vol. 51, No. 2, 299-320.

89

90

Researching ICTs in Context

Himanen, P. 2001. The Hacker Ethic and the Spirit of the Information Age. New York: Random House. Hine, C. 2000. Virtual Ethnography. London: Sage. Knorr-Cetina, K. 1999. Epistemic Cultures, Cambridge: Harvard University Press. Knorr-Cetina, K, & Cicourel, A. 1981. Advances in Social Theory and Methodology: toward an Integration of Micro- and Macro-Sociologies. Boston: Routledge & Kegan Paul. Latour, B. 1999. Pandora's Hope. Cambridge: Harvard University Press. Law, J. 1994. Organising Modernity. Oxford: Blackwell. Locke, C. et al. 2000. Cluetrain Manifesto. Cambridge: Perseus Books. Locke, J. L. 1998. Why We Don't Talk to Each Other Anymore: the de-voicing of society. New York: Touchstone. MacIver, R. M. 1931. Society: its structure and changes. New York: The Branwell Press. Merleau-Ponty, M. 1962. Phenomenology of Perception. London: Routledge. Miller, D. & Slater, D. 2000. The Internet: an ethnographic approach. Oxford: Berg. Rheingold, H. 1993. The Virtual Community. Reading: Addison Wesley. Robertson, R. 1992. Globalisation. London: Sage. Saxenian, A. 1994. Regional Advantage. Cambridge, MA: Harvard University Press. Sennet, R. 1998. The Corrosion of Character. New York: Norton. Sennet, R. 1976. The Fall of Public Man. New York: Knopf. Stehr, N. 1994. Knowledge Societies. London: Sage. Storper, M. & Walker, R. 1989. The Capitalist Imperative. London: Blackwell. Suchman, L. 1987. Plans and Situated Actions. Cambridge: Cambridge University Press. Sørensen, K. 1998. The Spectre of Participation. Oslo: Scandinavian University Press. Turkle, S. 1999. 'Looking toward cyberspace: beyond grounded sociology. Cyberspace and identity'. Contemporary Sociology ,Vol. 28, No. 6. 643-667. Undheim, T. A. 2002, under review. What the Net Can't Do. Ph.D dissertation. Faculty of Social Sciences and Technology Management, Trondheim: NTNU. Undheim, T. A. 2000. 'Getting connected: how sociologists can access the high tech élite'. STS Working Paper, 3.

Locating the Internet

Urry, J. 2000. Sociology Beyond Societies. London: Routledge. Wellman, B. & Hampton, K. 1999. 'Living networked on and offline'. Contemporary Sociology, Vol. 28, No. 6. 648-654. Wellman, B. et al. 1996. 'Computer networks as social networks: collaborative work, telework and virtual community'. Annual Review of Sociology, Vol. 22, No. 2. 13-238. Wenger, E. 1998. Communities of Practice: learning, meaning, and identity. Cambridge: Cambridge University Press.

91

92

Researching ICTs in Context

BLANK

5 Beyond the enigmatic utopia:

researching facts and failures in ICT projects1

Jarle Brosveet

Introduction More often than not, when Information and Communications Technology (ICT) systems are completed, they turn out to be deficient in one way or another. It is reasonable to believe that the traditional textbook models of performing ICT development work or systems engineering,2 as adhered to by most computer scientists, tend to produce unsatisfactory results for users and developers alike, not least because these models are far too simplistic, mechanistic and technology-driven, at least as viewed from a social science perspective.3 Since ICT work is being performed more or less jointly by users and developers, neither party is likely to admit that they are responsible for the shortcomings or, indeed, what is the nature of these shortcomings. The researcher dealing with the history of any ICT development project is likely to find that various claims are stated or refuted by those involved. Also, claims seem to change continually as the researcher confronts the parties with his findings. To the bewildered researcher, facts appear to be negotiable, even a long time after the ICT development project has ended. Characteristics such as these were expounded at great length, using a wide range of examples, by Bruno Latour when he launched the social constructivist method available to social scientists who attempt to sort out how facts come about (Latour 1987). Latour recommends that the researcher should follow the actors through various steps known as the translation process. Initially, the translation process was thought to involve the operation of a scientific laboratory in the interest of eliminating undue influences 93

94

Researching ICTs in Context

and enhancing the power of the key actor (Latour 1983). However, the concept of a translation process has since been extended so as to analyze events in society at large, with due regard given to extraneous influences even though this might render the analysis less poignant or less conclusive.4 In this paper, I will look at three ICT development projects, in the form of cases, in which facts seemed highly negotiable: • the municipal service bureaux case • the Fredrikstad electronic highway case • the medical ICT case. The actual outcome of these three projects will be contrasted with the strategies of users and developers as the projects were evolving. Since these strategies are often in conflict, the projects can be documented to produce results which are painful or embarrassing to users and developers alike. However, these failures are often not admitted by the participants. Instead, the researcher is presented with wishful thinking which tends to conceal that goals have not been met and that conflicts have not been resolved. Here I will apply a social constructivist perspective, especially by considering the variety of role-definition and scenario-building strategies found as part of the translation process in order for some of the actors to achieve dominance – both in the projects as they were developing and when the researcher attempted to establish knowledge of what had taken place.

The Municipal Service Bureaux Case5 In Norway, the seven regional municipal service bureaux served from the 1950s until the 1990s as organisations taking care of what in present-day parlance is known as the outsourcing of data processing. At the time, few municipalities did their own data processing, so each of the service bureaux acted as jointly owned enterprises for the local authorities of a certain geographic region of Norway. When the background and history of the municipal service bureaux were researched in the early 1990s, the bureaux were nearing the end of their life span. What had seemed a viable outsourcing concept in the 1950s had gradually eroded with the introduction of new technology and changes in user sentiment. When the research project was started, conflicts were all too evident between the technology-driven bureaux and their more result-oriented user-owners. The conflict escalated rapidly,

Beyond the enigmatic utopia

so before the research project was completed, the disillusioned and enraged municipalities had sold out to IBM and started using PC-based local area networks or made deals with new outsourcing partners. User reactions were obvious since they were told at great length in newspapers and magazines, so the research project concentrated on charting the fate of the defunct technology-driven bureaux, which no doubt had failed to adjust properly to the demands and expectations of their userowners. Initially, interviews were conducted with some of the old managing directors of the bureaux, now largely retired. It was expected that these interviews would provide insight into the prolonged decline of the service bureaux and make up for the lack of written documentation about their fate. It turned out that these interviews did not go as planned. Some of the retired managing directors were unwilling to talk, and those who did, obviously had decided to improve on the miserable reputation they left behind. In the interviews they tried very hard to minimise any suspicions of their own failure by accusing the user-owners of being an uncooperative lot. Although retired, the managing directors immediately took up the old battle axe and acted as if the interviewer were an investigator dispatched by one of their previous user-owners. Happily, a couple of archives were retrieved which helped enormously in establishing a more truthful picture of what had been going on in the declining years. In fact, a critical report commissioned by the bureaux as early as 1982 had warned about their likely fate. Also, many government reports had suggested that central government should take on some of the services run by the bureaux, in order to integrate ICT development with requirements posed by an expanding welfare state, especially in the field of tax collection and improved social services. The interviews revealed nothing about the existence of the critical 1982 report and did not inform about most government initiatives vis-à-vis the bureaux.

The Fredrikstad electronic highway case6 In June 1997, a feasibility study was completed for a projected information highway infrastructure — Frihus 2000 — in the municipality of Fredrikstad in south-eastern Norway. In this particular case, the information highway project was believed to contribute to increased industrial activity in an economically backward region. Also, plans were made for the information highway to attain a high degree of user-friendliness.

95

96

Researching ICTs in Context

This local initiative seemed a promising field of study, not least because it relied on a high degree of user participation. In this way Frihus 2000 proved very different from most other digital city projects studied within the scope of the European research programme of which it was a part.7 Also, finding such a project in an economically backward region was something of a novelty. When research started, five local user groups had just completed their requirement specifications. These proved much too incoherent and inconclusive for the local bureaucrats to chart a possible course of action. An interview showed that the local decision-makers had arrived at an impasse and were about to enlist the assistance of Telenor, the major telecom provider, in order to make further advances. In May 1999, newspapers reported that the electronic highway project had been terminated rather abruptly, and that results envisaged by the user groups as well as the notion of an electronic highway had vanished completely. The Telenor project leader was not available for an interview and representatives of the local authority were at first rather circumspect in their comments. Only one year later, the local authority's director of EDP finally admitted in an interview that the project was a failure, and that Telenor did not measure up to expectations. Again, there were accusations that the project had become too technology-driven and that user aspects had been overlooked despite efforts made initially by the user groups, which received scant attention by Telenor. Apparently, this conclusion was arrived at after a somewhat prolonged soul-searching process within the ranks of the local authority. The outcome, yet again, was to find the technocrats blameworthy, even though project development depended on the joint effort of developers and users.

The medical ICT case The medical ICT project, begun in 2000, was looked upon as an opportunity to follow the actions and interactions of developers and users continuously over a period of three years. In this manner it was hoped to uncover some of the finer points which had proved elusive when development is being analysed ex post facto by means of interviews and available documents. If things went well, an insider view might result which could be reminiscent of the classic laboratory study by Latour & Woolgar (1983/1986), although located outside an academic laboratory setting. In

Beyond the enigmatic utopia

particular, a parallell study of the actions of developers and users presented itself as a tantalising idea. These hopes were soon dashed despite a promising start. Aparently, agreement had been reached to follow the ORBIT project at the Norwegian Centre for Medical Informatics (KITH), a project which aimed at introducing electronic routines to serve doctors as well as a hospital in middle Norway. However, at the first meeting, after having presented his plans to conduct a laboratory study, the researcher was thrown out by the project manager, an engineer who resented the thought of being followed closely by an inquisitive and potentially critical social scientist. Unfortunately, the managing director (who later resigned) sided with the project manager so as to effectively block access to the ongoing deliberations between developers and users in the ORBIT project. Next, an attempt was made to follow the KVALIS project, in which students and staff in computer science at the Norwegian University of Science and Technology interacted with medical personnel in developing electronic patient journals for use in Norwegian hospitals. Again, it was made plain that the objective of such participation would be to observe the action and interaction of developers and users, and again the project manager firmly declined the wish of a social scientist to nose around, that is one whom he did not regard as a true follower of systems engineering textbook models to which his students were supposed to adhere. A third attempt was made to get a foothold inside an environment consisting of users as well as of developers of ICT systems within medicine. Contact was made with an assistant professor in social medicine and general practice who had worked incessantly during the 1990s to set up a national patient statistics register beneficial to the planning of medical services, in particular within primary health care. However, his valiant endeavour proved futile, and the task of maintaining the patient statistics register on a permanent basis was assigned to the Central Bureau of Statistics in spring 2000. Later, the same assistant professor was commissioned to do a feasibility study on the possible establishment of a national centre for the development of electronic patient journals. Obviously, such a centre would be an ideal place to study the interaction between users and developers should the management be sympathetic to the cause of social scientists. However, this centre, should it come about, would not be established within the three years set aside for the medical ICT project intended to analyse the actions of developers and users.

97

98

Researching ICTs in Context

At long last, it became clear that the medical ICT project could not be made into a laboratory study in the tradition of Latour and Woolgar. As a result, and once again, knowledge of the relationship between developers and users would have to be delineated ex post facto, mostly from available reports and supplemented by the occasional interview.

Who's afraid of the big bad wolf? In the three research projects outlined above, attempts were made to separate facts from fiction by approaching the cases from various points of view. In the municipal service bureaux case, the developers were followed closely, whereas in the Fredrikstad electronic highway case, events were described from a user's point of view. In the medical ICT case, plans were made to follow the actions of developers and users in parallell. To various degrees, all attempts at pursuing certain actor groups were frustrated. In the municipal service bureaux case, developers were not entirely honest about describing past events. Since the research project covered the story of a business concept going back over 50 years, one might think that forgetfulness played a part in the interviews. However, since vital elements of the story were found to be missing from the interviews, there is more than a slight suspicion that the developers' story was not entirely truthful. In a more forgiving view one might say that they looked upon it as their version of truth. Unfortunately, this version happened to be in conflict with the researcher's perception of truth after the developers' story was checked against contemporary documents. In the Fredrikstad information highway case, the Telenor developers ended up by being described as bad guys in the research report. In fact, their actions were likened to those of the pied piper of Hamelin. This coincided with the conclusion that the users arrived at eventually. Certainly, it must have pleased the users that the researcher's conclusion agreed with their fabrication of truth. Initially, it was not at all obvious that the developers would be to blame in the final report, because there had been an interplay between users and developers all along. The users had at any time the possibility to instruct or correct the developers, so the miserable outcome of the information highway project might just as well be the users' fault. If fact, the research report points out this ambiguity. The opposite conclusion might have been equally plausible if emphasis had been place:

Beyond the enigmatic utopia

… a gross oversight verging on blindness [which] afflicted the non-expert users who let themselves be taken into the proverbial mountain, which they undoubtedly must have seen or suspected. (Brosveet, forthcoming).

In the medical ICT case, it once more seemed as if the developers were to blame, as they prevented research into the relationship between developers and users on two occasions. No doubt, they saw the writing on the wall and tried to prevent yet another report with a user bias which would be embarrassing to the developers by giving them no credit or, even worse, discredit them. By denying the social scientist access they ensured that the developers' report would stand a better chance of defining truth uncontested. This move was perhaps not unexpected, considering the many past reports which are critical of the application of traditional systems engineering procedures when dealing with user problems. However, still more disturbing was the silence on the part of the group of users who in one of these cases witnessed that the project manager dismissed the social researcher in their presence. In fact, they did not seem to mind. When these three cases are scrutinised for similarities, the developers constantly emerge as the 'big bad wolf' callous to the cause of users not least because developers suffer from an unstinting reliance on mechanistic models inherent in their systems engineering background. However, this does not mean that the users can be said to conform to the image of the suffering 'three little pigs'. When the 'big bad wolf huffs and puffs to blow the house down', which is often the image evoked when computer systems are malfunctioning, it might as well be that the building of 'a house of straw' on the part the users is equally to blame.

Great expectations Considering our cases once more, it appears that the user-owners of the municipality service bureaux case were unable to lead their enterprises in a successful direction despite the power that they possessed. In the Fredrikstad information highway case, the five user committees made for an auspicious start but became invisible at a later date. In the medical ICT case, there was no apparent user initiative to secure the participation of a social scientist, which might have been in the users' interest. This defeatist stance is remarkable in the face of the issues at stake, because in each of these cases the users expected a positive result in accordance with their expectations.

99

100

Researching ICTs in Context

One might be forgiven for thinking that developers and users are living in separate worlds, which, to paraphrase Kipling, shall never meet.8 This seems obvious even if developers recognise the problems they have in meeting user needs and are making attempts to bridge the gap between these two worlds. In most cases the recommended solution consists of applying traditional engineering methods such as continuous quality control and improved timelines to the structuring of situations characterised by social expectations or dreams which cannot be easily captured. In applying these methods, clarity is the condition called for, such as clarity of perspectives, clarity of intrinsic and extrinsic values, and clarity of quality. Even so, the problems remain in systems engineering of finding proper ways to handle changing user requirements.9 One of the most vexing questions for developers in dealing with user requirements seems to be coping with the learning process, or, in their own phraseology, the re-usability of meta-architectural knowledge. In short, developers are often at a loss to determine what knowledge is applicable to new situations, and how re-usable knowledge can be derived from what they experience. Tentatively, and in the fashion of traditional engineering, the sorting out of user ideas and dreams has been suggested to require at least four categories as detailed in Table 1. Simple

Tacit and no structure Knowledge and structure

User requirements Complex

INTUITIVE Requires experienced developer to make up structure

EVOLUTIONARY Developers clarify nature of future development in processes of learning and 'unlearning'

PROFESSIONAL SYSTEMATIC Explicit and Traditional 'water- Knowledge is adequate, well struc- fall' model with clear so outcome depends on user specifications communication between tured or users and developers from start on which codified developers depend

Figure 1: User requirements vs knowledge and structure in systems development.10

Beyond the enigmatic utopia

As can be seen from Figure 1, which represents a computer science view, the existence of tacit knowledge is taken somewhat casually to be synonymous with lack of structure. However, this problem is believed to be soluble either by imposing a structure, if requirements are simple, or by going through heuristic learning or 'unlearning' processes, basically involving developers, if the requirements are complex. When knowledge is structured or codified, traditional systems engineering methods are looked upon as being sufficient if user requirements are simple. If not, the real difficulty lies in setting up efficient communication between developers and users. Regardless of the oversimplifications of Figure 1, it represents a radically new way of thinking about user expectations on the part of developers. This new line of thinking transcends the models of traditional systems engineering, which are classified as 'waterfall', i.e., divided into successive phases which are completed one by one, and which presuppose that user requirements are complete and unchanging from the outset. Still, there appears to be lacunae which computer scientists have not come round to trackling, basically because more radical approaches than those of establishing explicit knowledge and manageable structures deviate from engineering models to the extent that their introduction is frowned at in most instances as being far too outlandish. For this reason, grappling with tacit knowledge in other ways than imposing structure still awaits attention.11 That being said, computer scientists have long since come to realise that social and cultural contexts have to be taken into account. The sad thing is that, more often than not, computer scientists try to overcome the challenges of dealing with social and cultural contexts by a priori formulating simple rules and guidelines to be followed throughout the systems development process.12 This attitude conflicts with the notion found in the upper right hand corner of Figure 1, that the systems development process is in many instances basically a learning process – even to the extent of 'unlearning' preconceived facts – involving developers as well as users. Moreover, many computer scientists now ascribe to the notion that systems development, in order to be successful, will require the application of an improvisational change model.13 Clearly, something is astir in computer science, not least because there is now a faction who wants to beg, steal and borrow methods from sociology and ethnography in order to forge a multi-disciplinary

101

102

Researching ICTs in Context

approach to systems engineering.14 Notably, a new direction is emerging known as interpretive information systems research. This direction has developed out of the need for research in information systems to focus on social and cultural contexts, dealing with people's assumptions, beliefs and desires. In some cases, this direction is combined with action research in order for the researcher to be closely involved with the people and technologies that are being studied. In this way, research is being diverted from the theoretical simplifications and rationalistic bias of traditional systems engineering so as to achieve learning in the process of constructing improved information systems.15 As part of crafting new directions in systems engineering, theoretical underpinnings have been adopted in the shape of less mechanistic knowledge management models. Once a fuzzy and elusive concept, knowledge management has evolved into forming a basis for the understanding of how knowledge in collective memories is created. Also, knowledge management is conceived of as comprising ways of facilitating communication and co-ordination between people who actually create knowledge and those who need it. Based on this theoretical framework, procedures and guidelines have been developed for generating, storing, distributing, transferring and using knowledge in organisations.16 Again, the background is formed by the division between uncodified (tacit) knowledge and codified (explicit) knowledge. However, the guidelines sometimes verge on rigidity due to a lack of appropriate social and cultural contexts in which tacit and explicit knowledge operate. CODIFICATION STRATEGY Develop an electronic document system that codifies, disseminates and allows reuse of knowledge Invest heavily in ICT; the goal is to connect people with reusable codified knowledge Invest once in a knowledge asset; reuse it many times Reward people for using and contributing to document databases

PERSONALISATION STRATEGY Develop networks for linking people so that tacit knowledge is shared Invest moderately in IT; the goal is to facilitate conversations and the exchange of tacit knowledge Charge high fees for highly customised solutions to unique problems Reward people for directly sharing knowledge with others

Figure 2: Two strategies of knowledge management.

Beyond the enigmatic utopia

Bringing context into the picture, it has been argued that a uniform approach to managing knowledge is rarely taken and normally two different knowledge management strategies are employed. In some organisations the strategy centres on the computer and knowledge is carefully codified and stored in databases so as to be easily accessed and used by anyone in the organisation. This is called the codification strategy. In other organisations knowledge is closely tied to the person who developed it and is shared mainly through direct person-to-person contacts. This is called the personalisation strategy and uses computers mainly for communicating knowledge, not for storing it. Figure 2 presents the main differences between codification and personalisation strategies.17 Based on past experience it is hardly surprising that the codification strategy in knowledge management has received greatest attention due to the reliance on purely technological criteria. For instance, it has been said that the success of knowledge management tools built according to the codification strategy is dependent upon the efficiency of information retrieval.18 In this way, protagonists of this new direction are able to fall back on theories and procedures developed in the field of information retrieval a long time ago. Such falling back on the old and trusty obviously takes some of the sting out of the reorientation. While this helps to make the reorientation more palatable to a wider circle of developers, it also defuses the most radical aspects of this new direction and hampers the required reformulation of systems development tools and techniques.

Breaking illusions It appears that the incorporation of social and cultural contexts can prove a real obstacle when it comes to the reorientation of systems development in theory as well as in practice. As we have seen, amends are being made by introducing concepts such as learning and knowledge management, both of which involve aspects of fuzziness and heuristics not easily dealt with, most of all because these concepts, belonging to the behavioural sciences, are beyond the bounds of traditional computer science. As far as learning is concerned, several archetypical issues associated with social learning are in need of attention. At least three issues

103

104

Researching ICTs in Context

can be singled out which are at stake. These three issues can be regarded as central when trying to predict the outcome or even success of the systems engineering process:19 1. Ways of observing and characterising the efforts of designers in shaping technology and influencing human action in order to achieve control; 2. Ways of observing and characterising the material results of these efforts, such as the features of an actual ICT system; and, 3. Ways of assessing the efficiency of this effort of control of users of the ICT system under scrutiny. Economists and economic historians studying productivity in industry have been talking of phenomenons such as learning by doing and learning by using. Learning by doing focuses on the analysis of finding new or more efficient ways of using certain technologies. Learning by using focuses on the increased proficiency and skill of users in performing certain tasks. On this basis, a model of innovation has been formulated that takes into account learning by doing, learning by using as well as the introduction of a new dimension known as learning by making mistakes. Eventually, these three concepts have been merged into an approach called learning by interacting, which postulates that developers may depend critically on information from users, and vice versa, in order to innovate successfully.20 In addition to going more deeply into the questions of social learning, attempts should also be made to pay more attention to social conflict existing between and among developers and users. A case in point is this finding in a Danish ICT project that encountered a number of obstacles when performing a diagnostic analysis: One of these problem issues exposed an internal conflict within the company. This conflict was rooted in a dilemma about prioritising the IT solution platform. On the one hand to prioritise the IT solution as a generic system where new releases could be offered to all customers. On the other hand an approach where the individual customer's specific needs were prioritised in a way that could lead to different tailored systems, hard to maintain when new versions of the generic system modules became available.21

The advice in this particular case was to spend much time on constructive dialogue with all involved ICT consultants while stressing that the purpose was not to identify success stories or failures. Also, great care

Beyond the enigmatic utopia

was taken not to name individuals in reports and oral presentations. Apparently, social conflict was such a delicate matter that it could only be dealt with in a cursory manner. In particular, the manager did not appreciate that conflict was identified in the report. Furthermore, the manager suggested that his employees should not read the report. This example is instructive in pointing out the anxiety, at least among developers, of having social conflict identified and dealt with openly. Frequently, such conflict is suppressed or regarded as irrelevant so as not to be found worthy of attention. Again, traditional engineering attitudes intervene in a way that tends to block the road to adopting the kind of reformulation and rethinking that might help solve the more intricate problems existing between as well as among developers and users of ICT systems. Even so, actor-network theory has indeed been applied, using the devices of translation and enrollment as advocated by Latour (1987), to cope with situations in which actors – or actants, when including humans as well as non-humans – follow, twist or oppose inscriptions, which are programmes of action by users imposed by developers in the design of ICT systems. When a sufficient number of actors or actants are enrolled in the network, opposing interests may be forced to yield. As a result, a network may evolve towards a stable state with a more or less irreversible and unchangeable inscription, it may be in a state of flux and instability, or it may be only temporally stable.22 Actor-network theory is a bold way for systems engineers to proceed, not least because of the non-deterministic thinking lying at the base of this kind of theory. Also, emphasis within actor-network theory is on the interactive, continuous and open-ended nature of the design and implementation processes, implying a high degree of conflict resolution among actors and having inherent network instability as factors to contend with. This approach contrasts sharply with the traditional notion of systems design as being an utterly well-defined and conflict-free process of limited duration which leads to a stable and satisfactory result, technically as well as socially, provided that all relevant details are worked out carefully in the initial phases of problem formulation and design. A bewildering feature of actor-network theory is that it does not represent a unified body of concepts and that it can be used quite subjectively. Actor-network theory cannot be said to impart the same air of consistency and objectivity that surrounds many theories within sci-

105

106

Researching ICTs in Context

ence, even if there is enough evidence to say that science is more constructed and subjective than many would like to think.23 In fact, one might say that researchers as well as practitioners in science in general, and within computer science in particular, fall prey to an illusion of objectivity which does not exist. This kind of utopic thinking might explain why traditional theories and methods are adhered to, even when experience seems to indicate that social and cultural factors are at play which render these theories and methods useless in the case of complex user-technology issues such as those permeating systems engineering.

Pride and prejudice Although some younger computer scientists have come a long way recently in admitting that recognition of social and cultural contexts are pivotal in achieving good systems development results, there is still clearly room for improvement, particularly in the consultancy business where conventional and less cutting-edge models and methods are often relied on. It should be noted that much rethinking originates from university computer science departments and multi-disciplinary institutions in the late 1990s and early 2000s and that is only sparsely reflected in commonly accepted tools of the trade. For this reason, it is not surprising if some degree of suspicion or scepticism can be discerned both regarding the introduction of new methods and regarding research aimed at probing the depths of social and cultural awareness in systems design and implementation. Both aspects can evoke a sense of uneasiness because it conjures up a feeling of unwanted and undeserved criticism thought to be irrelevant under the circumstances. For instance, a Danish research report states the following: One consultant explained during the project, that they had already 'written' 80% of the final report for the customer even before they had the first visit at the customer site. This was immediately noted down by the researcher and later it turned out that the consultant had felt very annoyed by this. He was concerned how the researcher would interpret this 'work practice' and how it would be presented to other colleagues and managers.'24

In this case, the researchers engaged in introducing new design methods commented that either the developer was in fact a highly skilled individual who believed that most of his findings could be delineated from

Beyond the enigmatic utopia

past experience with other users, or he was just jumping to conclusions. However, the researchers did not look more closely into these alternatives for fear of compromising the main objective of their project, which was the introduction of new design methods. To the outside observer, it looks as if the developer did not want anybody to challenge the image of him being a highly skilled professional. He wanted the image of skilfulness established as an uncontested fact, so he felt ill at ease when a researcher took notice of his behaviour as being potentially condemnable. Apart from the psychological aspect of pride being hurt, there is also the sociological aspect of group dominance at work. Applying an actornetwork perspective, one might say that the Danish researchers succumbed to the lure of enrollment and network construction configured by the incumbent group of systems designers. By making clear to the researchers that they might jeopardise the main objective of their project if they were to pursue social and contextual issues, the systems designers protected their dominance successfully by defending the network of interrelationships they had established. In this manner the researchers were integrated as an unobtrusive part of the socio-technical ensemble represented by users, systems designers and researchers assisting in the introduction of new systems engineering methods. In action research, researchers always run the risk of being socialised or integrated into the group under scrutiny. The negative effects on researchers of such integration is that their critical stance is undermined to the extent that results reported are controlled or even dictated by the group being studied. If this happens, there is a risk that the researcher loses his or her objectivity, integrity and credibility (Argyris, Putnam & Smith 1985). On the other hand, resisting integration fully or in part can be a tough way to go, because there is always the possibility that those being researched upon will impose or threaten to impose sanctions of various kinds. The strongest sanction of all will be exclusion from further cooperation with the group, thus casting the whole research project into a shambles. Frustration is created by the researcher's wish to be critical and independent in a creative and positive manner when risk of exclusion is imminent. By being much to blunt in stating his or her intentions faceto-face with the group in question, participation in and insight into the inner workings of the group may be foreclosed. At the same time, studying the group from the outside does not present itself as a desir-

107

108

Researching ICTs in Context

able option, because participation is required in order to obtain elusive tacit knowledge residing in the group which is otherwise unobtainable. Undoubtedly, it is tempting for the researcher to enter into contact with the group being researched while putting on an air of innocence or pretence with the intention to fend off resentment or exclusion at an early stage. In this way, trust can be established which in many cases is the first and foremost condition to be met in securing access to the deliberations and inner workings of elusive social groups. Sometimes, the establishment of trust succeeds admirably and the researcher can triumphantly report that those being studied 'entrusted their stories to a stranger and a skeptic' (MacKenzie 1990: xi).25 Obviously, establishing trust under false pretensions is always morally and ethically doubtful, even if it is tempting enough so as not be to totally unavoidable. No doubt the cynical researcher can go a long way towards being hypocritical and still escape criticism at a later stage by arguing that his or her critical and sometimes ruthless presentation in the shape of factual results came as an afterthought. However, the question remains how much openness and sincerity is advisable in the early phases of research projects dealing with lack of social and contextual awareness of the kind found in many systems engineering projects.

Reappraisal Going back to the three cases presented initially, we may now reconsider some of the critical elements observed when a researcher gets involved with ICT developers and users as part of a social science research project. Evidently, the cases reflect many of the features already discussed, although to various degrees, both regarding the attitudes and outlook of developers and users, and regarding the intervention of a social scientist who no doubt was looked upon as some kind of intruder. In the municipal service bureaux case, the nature of developer-user conflict proved at first hard to describe. Developer and user descriptions and explanations turned out to be as propagandistic as those of any wartime foes. In interviews it was felt that the old bureaux managers entrenched whenever they could and spoke in a way that made truth seem highly negotiable. Any twist and turn seem permissible for as long as it served to elevate the noble cause of service bureaux developers, who had to cope, as it were, with an unruly lot of fickle users not knowing their own good.

Beyond the enigmatic utopia

In order to break the ice, the researcher had to explain to the old managing directors that the purpose of his work was to describe the history of the service bureaux, not to analyse conflicts and unsuccessful developeruser adjustment. It was important to stress that the researcher was not a journalist aiming for sensationalism, and that none of the material gathered would ever reach the newspapers. Still, many managers did not want to cooperate, and for a long time it seemed as if interviews were unobtainable. It can be argued that the researcher was not entirely truthful in telling the managers about his intentions. The purpose of the research right from the outset was to analyse as much as could be revealed of userdeveloper conflicts. To this end, actor-network theory would be applied to chart the scenarios, enrollment and inscriptions which defined developers as key actors and reduced users to play the part of squabbling and unsuccessful opponents. However, explaining these intentions to the managers would most certainly provoke an attitude of non-cooperation. At the time, it was deemed necessary to ferret out developer views and strategies through interviews in order for the project to succeed. For this reason the true intentions of the research project could not be disclosed. In the Fredrikstad municipal electronic highway case, the conception of developers and users was in a state of flux as the research project started. The electronic highway plans had reached a deadlock because five user groups had completed requirement specifications which, viewed as a whole, proved inconclusive. Up to this point, civil servants had served as developers, communicating with groups of potential users. Now a decision was taken to continue in a more technological direction by engaging telecom provider Telenor as developer. In this way the local authority relegated itself to a user role, and the five groups of real users vanished from sight. Again, the researcher stated that his intentions were to look at the history of the municipal electronic highway plans without going into the shift of focus regarding developers and users which, to be true, proved the really interesting point in this development project. A joint interview was arranged with two municipality officers which took an unfortunate turn because it exposed the lack of planning ability of the local authority. In fact, the electronic highway plans were exposed as being far too utopic to be handled successfully. The unprecedented reliance on user group requirement specifications was not the inspired initiative it looked initially. The interview revealed that it was an emergency measure resorted to by the local authority due to the inability to act on its own.

109

110

Researching ICTs in Context

Due to the utopic and sparsely formulated goals of this electronic highway project, one year passed before renewed contact was made by the researcher. Now, instead of conducting interviews, a press conference was taped at which presentations were made of the electronic highway plans, which by now had been completely revised. At this point in time, Telenor's presentation abounded with intricate high-tech visions, which contrasted with the municipality's presentation, which had reached a new height of utopic reasoning. A clear picture of developer-user conflict had emerged, and the researcher decided to wait another couple of months without making more interviews. Before long, newspapers announced that the electronic highway plans had been cancelled, and the researcher's report on the killing off of an electronic highway project could be written retrospectively (Brosveet forthcoming). To safeguard against a completely misguided analysis, an interview with a central municipality spokesperson was conducted before publication of the research report. This spokesperson did not disagree with the analysis, and for the first time it was revealed that the municipality considered the electronic highway project a failure from the day they judged the five user group requirements reports to be inconclusive. For the municipality, what happened afterwards served mostly to deflect criticism. They never considered themselves users after the real user groups were abolished, so Telenor's developers were in fact left in a void in their vain attempts to salvage the project, thus becoming the target of local criticism. In the medical ICT case, the researcher made attempts to join two on-going projects which were in the early stages of developer-user contact. Since the projects in question were to some extent rooted in a university environment while involving medical practitioners, the researcher expected a more sympathetic attitude towards his scientific endeavours than had been the case in the municipality service bureaux and in the electronic highway project. As explained above, this was not to be, as participation was denied effectively before the researcher had a chance to make his intentions clear. The two medical ICT projects had no long prehistory, so there was no way of trying to convince developers and users that establishing some kind of simple record of events would be the main focus of the intruding researcher. From past experience with social researchers, developers as well as users no doubt suspected that a potentially embarrassing critical report would result, which they could not prevent if the researcher was allowed

Beyond the enigmatic utopia

admission to the meetings so as to observe or take part in typical scenariobuilding, enrollment and inscription processes. Apart from the medical profession, participation would be restricted to people with the required systems engineering background, which evidently had to be of the traditional kind. In these milieus, the more radical rethinking of social and contextual concerns evidently had no place, or, if it had, was not assigned pride of place in the early stages of systems development.

Conclusions Trying to gain insight into projects which are more or less open to inspection can be a rewarding as well as a frustrating task, in particular when the purpose is to track down and analyse the machinations of developers and users in their network-building activities. It seems that the accessibility of projects as far as inspection is concerned, depends to some extent on the stage at which the social researcher approaches the actors. If the actors are in the early phases of problem formulation and scenario building, like in the two medical ICT projects, developers and users might feel so invulnerable as to reject any proposal of a social scientist to take part in the proceedings. Users Active

Inactive

Active

Medical ICT case

Fredrikstad electronic highway project (late phase)

Inactive

Fredrikstad electronic highway project (early phase)

Developers Municipal service bureaux Fredrikstad electronic highway project (abandoned)

Figure 3: Project activity levels of users and developers.

111

112

Researching ICTs in Context

When network building is further advanced, such as in the Fredrikstad electronic highway case, the gaining of access is less difficult, although it is questionable whether interviews will yield much insight during actual scenario-building. This is so irrespective of whether developers or users have the upper hand. Access by the social researcher is easiest in the final phase of a project or after the project has been completed, such as in the municipal service bureaux case. In this case, actors tended to have less vested interest in the project, as inscriptions had been completed that could be documented in various ways, so concealment was to some extent useless when confronted with the researcher's wish to conduct interviews. However, as we have seen, actors may still try to manipulate facts in a last attempt to gain a more advantageous reputation than would otherwise be possible. In Figure 3 the projects have been categorised according to the activity levels of developers and users. It appears that the attitudes of developers and users towards cooperating with the social researcher in shedding light on enrolment and inscription processes depend on the combined activity levels of these two groups. As a corollary, the researcher should be aware of his or her chances of succeeding by considering the developer-user activity levels and chose an appropriate course of action accordingly. These cases have shown that the possibility of success is greatest in cases where developers and users alike are presently inactive. In all other instances, developers as well as users tend to be weary of the intruding social scientist so as to render interviews ineffective or even refuse to be cooperative in providing insight into the social and contextual implications of the systems engineering process.

Refrences Aanestad, M. & Hanseth, O. 2000. 'Implementing open network technologies in complex work practices: a case from telemedicine'. In Baskerville, R., Stage R. & DeGross, J.I. (eds). Organisational and Social Perspectives on Information Technology: IFIP TC8 WG8.2 International Working Conference on the Social and Organisational Perspective on Research and Practice in Information Technology, June 9-11, 2000, Aalborg, Denmark. Boston: Kluwer Academic. 355-369.

Beyond the enigmatic utopia

Akrich, M. 1992. 'The de-scription of technical objects'. In Bijker, W. & Law, J. (eds) Shaping Technology-Building Society: Studies in Sociotechnical Change. Cambridge, MA: The MIT Press. Andersen, E. S. 1989. Systemutvikling. Bekkestua: NKI. Andersen, E.S. & Lundvall, B.-A. 1988. 'Small national systems of innovation facing technological revolutions: an analytical framework'. In Freeman, C. & Lundwall, B.-A. (eds) Small Countries Facing the Technological Revolution. London: Pinter. 9-36. Argyris, C., Putnam, R. & McLain Smith, D. 1985. Action Science. San Francisco: Jossey-Bass. Arrow, K. 1962. 'The economic implications of learning by doing'. Review of Economic Studies, 29. 155-173. Berg, Marc. 1998. 'The politics of technology: on bringing social theory into theoretical design'. Science, Technology, & Human Values, Vol. 23, No.4. 456-490. Bergenstjerna, Met al. 1999. Metoder för strategisk IT-management. Gothenberg: Department of Informatics. (Master's thesis) Beyer, H. & Holtzblatt, K. 1998. Contextual Design: Defining CustomerCentered Systems. San Francisco: Morgan Kaufmann. Bjerknes, G. et al. 1990. (eds) Organisational Competence in System Development: A Scandinavian Contribution. Lund: Studentlitteratur. Bødker, K., Kensing F. & Simonsen, J. 2000. 'Changing work practices in design'. In Svensson L. et al. (eds) Proceedings of the 23rd Information Systems Research Seminar in Scandinavia. Uddevalla: Laboratory for Interaction Technology. Brosveet, J. 1996. EDB inn i kommunene: Kommunedatatanken i aktørnettverksperspektiv. STS Report No. 26. Trondheim: Centre for Technology and Society. (Dr.polit. thesis). Brosveet, J. 1999. IBM Salesman Meets Norwegian Tax Collector: Computer Entrepreneurs in the Making. IEEE Annals of the History of Computing, Vol. 21, No. 2. 5-13. Brosveet, J. Forthcoming. 'Translation terrain and pied piper detours: how experts eliminated a norwegian it highway project'. Science, Technology, & Human Values. Button, G. & Sharrock, W. 1995. 'The mundane work of writing and reading computer programs'. In ten Havé, P. & Psathas, G. (eds) Situated Order: Studies in the Social Organisation of Talk and Embodied Activities. Washington: University Press of America.

113

114

Researching ICTs in Context

Dahlbom, B. & Mathiassen, L. 1993. Computers in Context: The Philosophy and Practice of Systems Design. Cambridge, MA: NCC Blackwell. Dahlbom, B. 1997. 'The new informatics'. Scandinavian Journal of Information Systems, Vol. 8, No. 2. 29-48. Dittrich, Y. 2000. 'Software engineering goes empirical: finding a path between understanding, intervention and method development'. In Dittrich Y. et al. (eds) Social Thinking – Software Practice: approaches relating software development, work, and organisational change. Wadem: IFIB. Floyd, C. et al. 1992. (eds) Software Development and Reality Construction. Berlin: Springer-Verlag. Hansen, M.T., Hohria , N. & Tierney, T. 1999. 'What's your strategy for managing knowledge?' Harvard Business Review, Vol. 77, No. 2. 106-116. Heathfield, H. & Loow, G. 1999. 'New challenges for clinical informatics: knowledge management'. Health Informatics Journal, Vol. 5, No. 2. 67-73. Hoffman, T. 1988. 'Corporate information systems strategy. In Pirow, P., Duffy N. & Ford, J. (eds) Information Systems in Practice and Theory. Amsterdam: Elsevier. Klein, H.K. & Myers, M.D. 1999. 'A set of principles for conducting and evaluating interpretive field studies in information systems'. MIS Quarterly, Vol. 23, No. 1. 67-93. Klein, H.K. & Kleinmann, D.L. 2002. 'The social construction of technology: structural considerations'. Science, Technology, & Human Values, Vol. 27, No. 1. 28-52. Kling, R. 1991. 'Computerisation and social transformations'. Science, Technology, & Human Values, Vol. 16, No. 3. 342-367. Latour, B. 1983. 'Give me a laboratory and I will raise the world'. In Knorr K. & Mulkay, M. (eds) Science Observed: Perspectives on the Social Study of Science. Thousand Oaks: Sage. Latour, B. 1987. Science in Action: How to Follow Scientists and Engineers Through Society. Milton Keynes/Cambridge, MA: Open University Press/Harvard University Press. Latour, B. 1992. 'Where are the missing masses: sociology of a few mundane artefacts'. In Bijker, W. & Law, J. (eds) Shaping Technology-Building Society: Studies in Sociotechnical Change. Cambridge, MA: The MIT Press. 225-259. Latour, B. 1999. Pandora's Hope: Essays on the Reality of Science Studies. Cambridge, MA/London: Harvard University Press. Latour, B. & Woolgar, S. 1983/1986. Laboratory Life. The Social Construction of Scientific Facts. Los Angeles/Princeton: Sage/Princeton University Press.

Beyond the enigmatic utopia

Lundeberg, M., Goldkuhl, G. & Nilsson, A. 1981. Information Systems Development: A Systematic Approach. Englewood Cliffs, NJ: Prentice-Hall. MacKenzie, D. 1990. Inventing Accuracy: A Historical Sociology of Nuclear Missile Guidance. Cambridge. MA: The MIT Press. Macredie, R.D. & Sandom, C. 1999. 'IT-enabled change: evaluating an improvisational perspective'. European Journal of Information Systems, Vol. 8, No.4. 247-259. Maidique, M.A. & Zirger, B.J. 1985. 'The new product learning cycle', Research Policy, 14. 299-313. Nilsson, A, G. et al. 1999. 'The business developer's toolbox: chains and alliances between established methods'. In Nilsson, A.G, Tolis, C. & Nellborn, C. (eds) Perspectives on Business Modelling: Understanding and Changing Organisations. Heidelberg: Springer-Verlag. Nonaka, I. & Takeuchi, H. 1995. The Knowledge-Creating Company. New York: Oxford University Press. Nygaard, K. 1986. 'Program development as a social activity. In Kugler, H.-J. (ed.) Information Processing 86: Proceedings of the IFIP 10th World Computer Congress, Dublin, Ireland, September 1-5, 1986. Amsterdam: North-Holland. Orlikowski, W. J. 1992. 'The duality of technology: rethinking the concept of technology in organisations'. Organisation Science, Vol. 3, No. 3. 398-427. Orlikowski, W.J. & Baroudi, J.J. 1991. 'studying information technology in organisations: research approaches and assumptions'. Information Systems Research, Vol. 2, No. 1. 1-28. Polanyi, M. 1967. The Tacit Dimension. London: Routledge & Kegan Paul. Rosenberg, N. 1992. Inside the Black Box: Technology and Economics. Cambridge: Cambridge University Press. Shapin, S. 1994. A Social History of Truth: Civility and Science in SevententhCentury England. Chicago: University of Chicago Press. Sørensen, K. H. 1996. 'Learning Technology, Constructing Culture: Perspectives on Socio-Technical Change'. STS Working Paper 18/1996. Trondheim: Centre for Technology and Society. Suchman, L. 1987. Plans and Situated Actions: the problem of human-machine communication. Cambridge: Cambridge University Press. van Bastelaer, B. & Lobet-Maris, C. 1999. (eds) Social Learning Regarding Multimedia Developments at a Local Level: the case of digital cities. Namur: CITA-FUNDP. Walsham, G. 1995. 'The emergence of interpretivism in IS research', European Journal of Information Systems, Vol. 4, No. 2. 74-81.

115

116

Researching ICTs in Context

Walsham, G. 1997. 'Actor-network theory and is research: current status and future prospects'. In Lee et al., Information Systems and Qualitative Research. London: Chapman & Hall. 325-343. Watne, K., Roos J. & von Grogh, G. 1996. 'Towards a theory of knowledge transfer in a cooperative context'. In von Grogh, G. & Roos, J. (eds) Managing Knowledge: Perspectives on Cooperation and Competition. Thousand Oaks: Sage. Woolgar, S. 1991. 'Configuring the user: the case of usability trials'. In Law, J. (ed.) A Sociology of Monsters: essays on power, technology and domination. London: Routledge. 58-99. Zuboff, S. 1988. In the Age of the Smart Machine: the future of work and power. New York: Basic Books.

Notes 1

This chapter was written as part of the research project 'Local networking cultures in Norwegian health services', financed by the SKIKT programme of the Norwegian Research Council. 2 In this report, ICT development work, systems engineering, systems development and systems design are taken to mean the same thing. Also, the terms systems designers and developers are used interchangeably. 3 Simplistic and mechanistic tendencies, most of all evident in rationalistic Anglo-American models of systems engineering, are modified to some extent by the introduction of aspects of organisation and workplace considerations as well as aspects of participatory design and action research. These are characteristics of what has been termed the soft systems approach or socio-technical approach to systems engineering, much favoured at universities throughout Scandinavia. To some extent, this approach to systems engineering has existed since the 1960s and has been propagated by academics such as Nygaard (1986) and Andersen (1989) in Norway, Lundeberg, Goldkuhl and Nilsson (1981) as well as Dahlbom (1997) in Sweden, and Mathiassen (Dahlbom & Mathiassen 1993) in Denmark. Another way of circumventing the mechanistic rigidities is represented by the field of computer-supported co-operative work (CSCW), as explained by Bjerknes et al. (1990). One of the most explicit international critics from a social science point of view of rigidity in computer science is Kling (1991). Floyd (1992) is one of the most explicit critics from the ranks of computer scientists, arguing that software development is social construction. 4 Latour (1999) has contributed to this trend by publishing essays on the reality of science studies, while keeping the analysis focused on the fabrication of academic knowledge. Others have been more intent on breaking out of the

Beyond the enigmatic utopia

laboratory setting in order to include limitations imposed on the translation process by the real world, e.g., Berg (1998); Klein & Kleinman (2002). 5 The case was analysed in full in Brosveet (1996). An analysis of hybrid roles that developed as the Bureaux were being established can be found in Brosveet (1999). 6 See Brosveet (forthcoming) for a full report. 7 The research was undertaken within the MULBRU project, as well as within the Social Learning in Multimedia (SLIM) project, supported by a grant from the European Commission (DG XII) as part of their Targeted Socio-Economic Research (TSR) programme. A summary report is provided by van Bastelaer and Lobet-Maris (1999). 8 'Oh, East is East, and West is West, and never the twain shall meet, Till Earth and Sky stand presently at God’s great Judgement Seat.' From Rudyard Kipling's 'The Ballad of East and West', 1892. 9 Nilsson et al. (1997) has addressed the developers' problems of coping with the ever-changing requirements of a coordinated enterprise-wide systems development process. Many suggested improvements in dealing with social conflicts by means of a meta-architectural approach to systems engineering is reviewed by Bergenstjerna et al. (1999). The notion of applying a metaarchitectural approach originates from Hoffman (1988). Dahlbom (1997), in an attempt to define what he has termed New Informatics, is suggesting an alternative conceptual framework in which the goals or subject-matter fields within an enterprise are comparable to those of four different worlds. 10 Table 1 has been adapted, with additional comments, from Bergenstjerna et al. (1999). 11 A model for converting tacit knowledge into explicit knowledge by socialisation and for converting explicit knowledge into tacit knowledge by internalisation is presented by Nonaka & Takeuchi (1995). They draw on the work of Polanyi (1967) in distinguishing between these two types of knowledge. 12 A recent example is Beyer & Holtzblatt (1998), who specify five different models for analyzing context: flow model, sequence model, artifact model, cultural model, and physical model. 13 Orlikowski (1992) has noted that similar technologies implemented in different contexts seem to have very different organisational outcomes. Macredie & Sandom (1999), using a different case, prove the relevance of an improvisational change model advocated by Orlikowski. Also, Zuboff (1988) has pointed out that ICT projects not only automate but also informate, i.e., produce new knowledge, a fact which can have implications for different work practices. Suchman (1987) describes how an expert system lacks access to contextual and social resources that humans normally use in interaction, and thereby fail to support various user requirements. 14 At the International Conference of Software Engineering 2000, a workshop was arranged titled 'Beg, borrow or steal: using multi-disciplinary approaches

117

118

Researching ICTs in Context

in empirical software engineering research'. 23 papers were presented. Also, the lack of influence from other disciplines has been pointed out by Button & Sharrock (1995) and Dittrich (2000). An appraisal of the influence of actornetwork theory on ICT research is found in Walsham (1997). 15 Interpretive information systems research is surveyed by Walsham (1995), with reference in particular to Orlikowski & Baroudi (1989). The notion is taken one step further by Klein & Myers (1999). A broad introduction to action research is given by Argyris, Putnam & Smith (1985). 16 Nonaka & Takeuchi (1995) and Wathne, Roos & von Grogh (1996) are central in defining the nature of knowledge management. 17 The notion of knowledge management contextualisation, as expressed in Table 2, originates from Hansen, Nohria & Tierney (1999). 18 First, the codification strategy is based on the idea that vital knowledge can be codified, stored and reused, which is the normal computer science way of thinking about problem-solving. Also, it gives an opportunity to achieve scale in knowledge reuse and thus of growing the business (Hansen, Nohria & Tierney 1999). Second, the efficiency of information retrieval techniques satisfies beliefs that time and effort are the major barriers to knowledge access and to achieving proper knowledge management (Healthfield & Louw 1999). 19 This point is made in more general terms, without mentioning systems engineering, by Sørensen (1996). Akrich (1993) and Latour (1993) have observed and characterised the efforts of designers in shaping technology and influencing human action in order to achieve control. The approach of these writers is to show that designers inscribe certain programmes of action by users in the design of a given artifact or technological system. In a similar vein, Woolgar (1991) has described software design as efforts to configure users. 20 The concept of learning by doing originates from Arrow (1962), and the concept of learning by using from Rosenberg (1982). The notion of learning by mistakes was added by Maidique & Zirger (1985). Linking users and developers to these concepts by introducing the idea of learning by interacting was made by Andersen & Lundwall (1988). It should be noted that these writers deal basically with product development, not with the development of ICT systems. 21 This is part of the reflections on activities in three Danish ICT organisations to change work practices in early design activities, as explained by Bødker, Kensing & Simonsen (2000). The activities were related to the introduction of new design methods based on theoretical studies as well as experiments. 22 Actor-network theory was used by Aanestad and Hanseth (2000) to describe how the interrelationship between work practice and technology unfolded as multimedia communication technology was introduced into an operating theatre. This case provides hints at relevant and potentially problematic issues to consider within telemedicine when a non-deterministic approach is chosen.

Beyond the enigmatic utopia

Also, the authors think that their study might influence the way that computer scientists think about design and implementation in general. 23 Many studies bear this out, e.g., the study by Shapin (1994) on civility and science in 17th century England which illustrates that science is indeed shaped by its cultural and social context. Social shaping of scientific facts also is the main interest of Latour (1987, 1999). 24 Experience made by Bødker, Kensing & Simonsen (2000) when dealing with changing work practices in early design activities within three Danish ICT organisations. 25 MacKenzie (1990) does not reveal how blunt he was in stating his real objectives, which were to describe in a critical way the development of nuclear missile guidance systems. Regarding the vital importance of tacit knowledge in completing his study he admits: The documents alone, however, would have made little sense to me. The most important input into the research came from people involved in the development of missile guidance and inertial navigation. I began by imagining that it would be hard to get anyone to talk to me. It was not. (MacKenzie 1990: x).

119

120

Researching ICTs in Context

BLANK

6 Experiments along the bazaar-route: the importance of user-producer dialogue in shaping new media technology Per Hetland

The Internet for all? Technological mediation involves representation and inscription in the world.1 One may therefore claim that engineers become sociologists through their development of hypotheses about what other people want and need. As a result, a large part of the work of innovators becomes the writing of user-values into the technical content of new objects and then the enrolling of other actors in support of these inscriptions (Callon 1987). These inscriptions, however, do not only apply to the artefacts as such but also to the situations in which they are used. In this way mediation translates and converts. I have earlier argued that science and technology mediation takes place in accordance with three different communication routes: (1) the direct route, (2) the middleman-route and (3) the bazaar-route (Hetland 2002). The direct route is associated with a clear definition of who is the sender and who is the public. It is the expert who popularises or problematises the scientific and technological knowledge for the general public. The middlemanroute is associated with a notion that the expert does not understand the media’s possibilities and limitations, or does not have time, or does not see the news value of the research material. Professional communicators therefore become necessary. The bazaar-route is an expression for a complex communication situation. What is relevant knowledge and how this is understood by different actors is the object of a dialogue.

121

122

Researching ICTs in Context

This dialogue is of special significance in what Callon calls hot situations (Callon 1998). In cold situations it is easy to identify actors, interests, preferences and responsibilities. One can therefore call in the experts and their laboratories. In hot situations most things are the subject of controversy, and those who are laymen want to have their say. These controversies are an expression of the fact that one does not have a stable basis of common knowledge and insight that one can agree on. Therefore technology producers, politicians and the media all see it as important to open up the way for feedback from the public in the diffusion process. The basis of common knowledge and insight that people formerly agreed on is constantly undergoing more frequent change. The establishing of common frames of reference is therefore a complicated process, particularly in hot situations. In such situations the way is also opened up for what Callon calls overflow. Overflow is an expression for the fact that complete framing is in many ways impossible. The degree of overflow is therefore an indication of how stable the frames are. Callon intimates two different ways of understanding the relationship between frames and overflow. In the first case establishing frames is the normal thing and overflow is leaks. In the second case overflow is the normal thing and the constituting of frames has high costs and will always be deficient. In this chapter I shall explore the bazaar-route with my point of departure in a social experiment that was going on at the same time as the Internet was being introduced to the public. I shall therefore look more closely at how a small group of users mobilised different arguments in the task of making their mark on the political agenda. The analysis takes as its starting point two important characteristics of the emergence of new media technology: 1) the highly developed capability of the modern economy to produce and diffuse user-values with new characteristics and 2) demonstration experiments as quasi-experiments2 creating an arena for interaction between users and producers. In demonstration experiments potential user needs for and potential user-values of new technology and new services are communicated between users and producers to facilitate re-invention and diffuse innovations. Demonstration experiments have often been used to promote special solutions among selected user groups to increase the general level of knowledge. They therefore represent a hybrid composition of technological and social innovation processes. In themselves they are excellent examples of the bazaar-route. To illustrate the methodological

Experiments along the bazaar-route

points I have used examples from one demonstration experiment in which I have participated: the adaptation of the Internet to the dissemination of public information to users with visual impairments. Politicians in large parts of the western world have put phenomena like 'the information gap’ or 'digital divide’ on the agenda. Enormous resources are to be used to remove the digital inequalities. In this work it is possible to identify different understandings of these digital inequalities. Three prominent ones are 1) access, 2) digital literacy and 3) confidence. With increasing access an interest has arisen in what is called digital literacy. This interest attempts to penetrate behind superficial views of the kind that 'the technology is so user-friendly that anybody can use it’. Any parallel comparison at all with general literacy, that is to say the degree to which the citizens are able to read and write, indicates that there are great differences here. I shall distinguish between two main forms of digital literacy: (1) tool literacy and (2) literacy of representation (Tyner 1998). Among literacy of representation one find visual literacy for understanding and using visual images in a communicative context. Automation has so far and to an increasing extent presupposed that the users act on the basis of visual impressions. In relation to the visually impaired, technological progress may therefore be experienced as the opposite. When it comes to technology policy one therefore observe a growing interest for universal design or 'Design-for-All’, as it is also called. This makes it important first to take a closer look at the importance of design in relation to the digital divide and thereby also digital literacy.

The importance of design In the infancy of computer age, only experts had access to this technology. With an extended user group, the desires for greater userfriendliness also arose. It will always be problematic to define userfriendliness, not least because the users constitute a multifarious group. Seen in a historical perspective, we can however link user-friendliness to the development of different interfaces. To study user interface we can start off with the hypothesis that machines display, at least in principle, interpretative flexibility. This opens the way for a study of the construction process (as a process of inscribing or writing) and of use (as a reading process). The relationship between the reader and the

123

124

Researching ICTs in Context

writer may be understood as being mediated by the machine and the interpretations of what the machine is, what it exists for, and what it can do (Woolgar 1991). Notions of future use and users of technology are in other words an important element in technological development. Suppositions about gender, competence, job performance and working environment in the relevant target group become an important part of technological design and become firmly anchored together with this. In other words, we meet technological objects not only as tools but also as bearers of meaning and of the interpretation of social and societal relations. These are properties that do not however prevent the users from re-interpreting and/or re-inventing these objects. It is therefore important to make technological problems comprehensible on the basis of the context in which the technology is placed. An obvious starting point in order to achieve this aim is to have a closer look at how technology constantly undergoes changes. The different strategies for interface design between users, the programme and the digital machine from menus to function keys and interactive graphic objects, reflect the degree of competence and control that one assumes the user has or must have. In 1981 Xerox launched its Star machine, which was a market fiasco but at the same time a model for the future user interface − the graphic user interface. Star was the first computer on the market to organise text production by means of pictures, or icons, in a visual language. Furthermore the machine made possible direct manipulation of graphic symbols by means of a mouse. The successor to Star, Apple, further developed the graphic user interface which became dominant for all machines/programmes in the course of the 1990s. This user interface was developed for the male knowledge worker (Hofmann 1995). The Star designer assumed that what the knowledge worker learnt about the use of computers was just as quickly forgotten as it had been learnt, because these people were not regular users of the technology. It was this large new group of part-time writers that now became the interesting target group for new wordprocessing programmes. The knowledge worker as a 'dilettante’ at the computer was now the user who was written into the programmes. The needs and the competence that the knowledge worker has, have since laid the foundation for a universal standard for user interfaces. Consciously or unconsciously, different schemes of events are therefore written into the technology. It is thus the social scientist’s task to

Experiments along the bazaar-route

decipher these scripts in order to increase our insight into why technology functions as it does. If you are among those who are reading this book in the form of black writing on white paper, then you are reading black writing. Black writing designates, among many who are visually impaired, the type of writing that sighted people read. As a metaphor it may just as well mean that so-called ordinary writing is inaccessible. Against this background this demonstration experiment, e-com, took its starting point in the situation that: (a) people with visual impairment today have no simple access to public information, either from the central or local authorities. This applied particularly to information about rights and obligations and (b) the public administrative authorities have only to a small extent seen to it that the visually impaired shall be able to communicate by means of new media technology (Hetland 1996 b). Eighty people who were visually impaired took part in e-com, and the information to which one had access was made available in three different forms (the proportion of users in e-com who read in the manner in question is mentioned in parentheses): (1) as magnified writing on the screen (31%), (2) as Braille on a reading line (53%), (3) as synthetic speech (16%). Some of the participants who used synthetic speech, used it to navigate around in the text or for rapid reading, but for closer reading they read the text by means of a reading line. Text can be written out in permanent writing − either as black writing (for those who can see) or in Braille (for those who are visually impaired). In addition Braille users can also be given fleeting presentations of Braille, corresponding to the presentations that sighted people get on the screen. The most common form of tactile presentation of this kind is an electronic reading line. Typically an electronic reading line for Braille consists of a single row of either 20, 40 or 80 sign-cells. Each sign-cell consists of a matrix of 4x2 points (4 vertically, 2 horizontally). Each point in the sign-cell is individually driven by a separate motor. In this way each individual sign-cell can generate any letter or figure whatsoever. The visually impaired often use great magnification on the screen and can take in five or six letters at a time. The blind use a reading line, but their fingers perceive only three or four letters at a time. People who have become visually impaired in adult life will as a rule have problems reading equally fast from a reading line as those who have learnt Braille as children. People therefore act within frameworks of action which are partly constituted by the media, at the same

125

126

Researching ICTs in Context

time as the media are not neutral tools; they colour and shape actions. Without falling into technology determinism one can thus state that it is interesting to study the coming into existence of technologies, because the resulting technology has significance and it is interesting to study its use and to concentrate on the variety of ways in which the same technology is used, since both technology and the way it is used have significance. In other words new technology has ambiguous consequences, and it can often be exploited in various ways with different results. Therefore it is necessary to analyse technology as a process in which a number of choices are made rather than to observe the choice that has been made when a given technology has been made available. The different problems the user faces are an important source for the description of how solutions are designed in relation to the users. Experiencing crisis also allows the setting to be described; 'if everything runs smoothly, even the very distinction between prescription and what the actor subscribes to is invisible because there is no gap, hence no crisis and no possible description’ (Akrich & Latour 1992: 261). Crises and problems are therefore our most important gateways to understanding what is happening. A blind person’s meeting with the automated post office illustrates this point: In the past you could simply queue up and you were served when it was your turn. Now you are marginalised by having first to find the machine that dispenses queue numbers and then press the button for a ticket. Then you must get somebody to read aloud for you what number you have been given and which number is now being served. You then try to count the pips until they reach your number, and if there is only one queue, you count correctly, but if, for example, there is one queue for ordinary services and one queue for foreign exchange, so that you cannot keep track of the pips, you must get somebody to tell you when it is your turn. If that is not enough, you must also get somebody to tell you which till you will be served at and then you must get somebody to take you to that till. At an early stage it had been thought that the queue number syst em could create problems for the blind. In an early system that was installed in some places in Sweden, when the customers came in, they broke the beam of light of a photo-electric cell. When the beam was interrupted, a machine with stored speech was actuated and it told the customer that this was a post office with a queue number system. The customer was referred to a service button on the queue number

Experiments along the bazaar-route

dispenser and if one pressed it, the number one had been given was read aloud. When it was the turn of the customer concerned, this number would automatically be read aloud over the loudspeaker system at the same time as a buzzer would be actuated at the till with the number in question. In this way the blind customer was guided to the correct till. The solution that was tried out in Sweden thus illustrates the fact that it was not the technology that made necessary the exclusion of the blind. The Norwegian version of the queue number system involves a number of changes. Among other things it involves a redefinition of what a queue jumper is. This is no longer a person who has been waiting for a shorter period of time than you have, but also a person who lacks a queue number and who either believes it is his/her turn or chooses to challenge the user representations that lie in the queue number system. Some blind people therefore choose to invoke their 'blindness’, or to protest against this 'blinding’, by going straight to the till and asking for service. Since those serving find it difficult to dismiss this strategy, these people will as a rule be served at once. The disadvantage of such a strategy is of course that the blind person is thus forced to confirm that s/he is disabled in relation to the user representation that lies in the queue number system. In other words the whole apparatus of technology and service producers right down to the end user contributes to a definition of the user and thereby also establishes parameters for the user’s actions. The special properties of a special screen image, or of the queue number system at a post office, must be understood on the basis of the cultural values and notions that precede design and the process of introduction − not as a reflection of specific technological possibilities. Script is therefore an important type of cognitive structure. Akrich and Latour see script as a series of actions inscribed in the technology by the engineer, inventor, manufacturer or designer (Akrich & Latour 1992). We have scripts for how plays are performed at the theatre, for ceremonies and also for daily phenomena such as visits to the post office. It is common for scripts to include remedies (queue numbers and screen images), roles (customer, fellow customer and person serving) and rules for the sequence of events (taking a queue number, waiting in the queue, going to the right till and being served). One of the most important sub-scripts that is placed in much of new technology is connected with the phenomenon of visualising. Here, one may also say that the script is con-

127

128

Researching ICTs in Context

nected with the general process of automation in which personal contact is replaced by machine contact. The sense of sight is thus made one of the technology developer’s most important frames of reference in development work; a high level of visual literacy is therefore made necessary. In what follows I shall therefore have a look at some of the problems that were registered and the solutions that were chosen when the presuppositions relating to visual literacy were present only to a limited extent.

Digital literacy translated Digital literacy is a question of a two-sided relationship. It is not simply the case that the users need a certain digital literacy in order to manage the technology at a certain level; it is also the case that particular technological solutions require a particular literacy. These requirements are often decided in the development and design process. Special user groups with physical or mental disabilities illustrate the problems that are often created in this connection. People have therefore been concerned with the fact that different user groups need technological solutions adapted to their needs. In practical policy this has manifested itself in an interest in universal design. We can say that a particular type of design presupposes particular literacy among the users; a given design in other words configures the users in a particular way. The purpose of universal design is to prevent unnecessary configuration of the users. The distinction between tactile mode and visual-symbolic mode is crucial for the visually impaired. In text mode the characters are transmitted to fixed positions on the screen (a specific number of characters per line, a specific number of lines per screen, for example). In graphic mode the information is transmitted to the screen as points. Therefore it is not possible simply to translate a picture of a character so that it can be presented in Braille and synthetic speech. Menus are a complicating factor for the visually impaired, not only because one must find one’s way about on the screen before one chooses anything, but also because attributes (colours and the like) are used to mark positions instead of the computer’s standard cursor. The cursor is the point on the screen at which one must direct one’s attention, a point of orientation. In such cases special equipment for Braille and synthetic speech will not simply go ahead and recognise what is the cursor. A window is a delimited field on the screen, often with a frame around it. Furthermore, it may overlap other windows. One window is the active

Experiments along the bazaar-route

window. When one can read several windows on the screen at the same time, it is difficult for the visually impaired to decide which window is the active one. People with weak sight get another type of problem; when the screen image is magnified, one sees only a small part of the screen. The purpose of e-com was to find out how public information could be made available to the visually impaired in the best possible way. A desired effect of e-com was that one should conserve and further develop suitable communication software for the visually impaired, who could not make use of the present graphically oriented user interface. In keeping with the development of new graphic user interfaces, a desire was formulated for continuous and goal-oriented development of software for the visually impaired. In the light of this e-com saw it as an objective to develop itself into a centre for electronic communication for the visually impaired. Further e-com saw it as an advantage not to use synthetic speech as the only medium, for two reasons. The systems that were used were not good enough (this applied in particular to languages of limited diffusion), and speech did not provide the opportunity to discover how words were spelt. It therefore became important to give the visually impaired the opportunity to maintain a written language of their own. Adaptation was therefore looked on as a means to reducing the information costs for groups, who on account of different factors had to bear far higher information costs than the expenses other actors experienced. On the other hand e-com emphasised that the aim of adaptation was to give the individual user possibilities of solving problems off her or his own bat. Adaptation was therefore an obligation that was partly incumbent on the sender − the message was to reach the whole target group − and it was an obligation that was partly incumbent on those who were responsible for different infrastructure measures − the infrastructure must not set up barriers to accessibility. E-com implemented a number of different measures to demonstrate the importance of adaptation, and here I shall only briefly present two of the most important ones: the newsletter and the local government elections in 1995. The newsletter was the original core of e-com. In the course of the project period (1995-1996), 33 issues were published − in the beginning one every fortnight − from the autumn of 1995 one per week. After the first trial period the newsletters contained on average 20 news reports. For all the newsletters the content consisted on average of news from state bodies

129

130

Researching ICTs in Context

(42%), local news from municipal bodies (15%), news from the Storting, or Norwegian national assembly, (11%), announcements (5%), and miscellaneous news (27%). The length of each news item varied somewhat. In an intermediate period the news items varied in length from 50 to 200 words, with the greater part being between 100 and 150 words. A midway evaluation elicited the response that many wanted greater variation in the length of the news items, which indicated that the user group was more heterogeneous than had originally been supposed. In the autumn of 1995, e-com started an additional project called Local Government Elections 1995. The point of departure was that the visually impaired had had limited opportunity to become involved in the elections in a 'democratic' manner. That is to say that they seldom received information about the names of the candidates on the list and they were seldom able to use their right to move candidates up the list or to delete candidates' names. Municipalities were chosen according to where interested users lived. The fact that the electoral lists were included meant that the names of 25 000 local government politicians were put on the net. Those who were used to reading by means of a reading line/synthetic speech had had few aids provided when it came to reading electoral lists and other material about the elections. Especially those who were completely blind therefore found this measure useful. The newsletter entailed specially adapted information. In this connection, two conflicting signals emerged. The one signal was that the visually impaired wanted to have access to the same information as normally sighted people. The other signal was that the visually impaired wanted to have an adapted text that was as effective as possible, so that they too, as rapidly and effectively as possible, could have access to the information they wanted. Once I have submitted a contribution to a discussion, I'm more or less obliged to follow the whole discussion, no matter how much rubbish there is in it. After all, it may happen that they attack me on some points, so that I must respond. It's just as difficult as keeping up with newspaper debates with readers’ contributions. And all this boils down to the fact that I, as a blind person, can't have an overview of such huge amounts of information as those who can see can have an overview of. The acquisition of information is slower. This is a limitation that the disability sets, whether you like it or not. So I think it's fine that the information, at least the information that is specially adapted for us, is concentrated. The importance of adaptation was documented in the course of the project period, and six examples were identified of how different scripts and

Experiments along the bazaar-route

frames of reference cause communication problems for the visually impaired: (1) The party one is communicating with does not understand what possibilities the visually impaired have to acquire relevant information. The information thus becomes inaccessible, not because the visually impaired person cannot acquire it, but because the person who is to communicate it believes it to be inaccessible. (2) The sender demonstrates lack of insight into the problem of accessibility. (3) The visually impaired are defined as a non-relevant social group. (4) The visually impaired are not considered as equal partners in the communication situation. (5) The sender lacks the necessary rights to make the information available. (6) The public room is privatised. In what follows I shall deal with an example of crises. Many of those taking part in e-com had experienced the sender’s problems in relating to the visually impaired as equal partners in the communication situation. One of the users contacted Vinmonopolet (the Norwegian state monopoly of wines and spirits) to obtain its price lists with information about the selection of wines and spirits offered, in electronic format. The first telephone call resulted in the answer that this was impossible. Since this was a bit difficult to understand, seeing that the lists certainly existed in electronic format, he pursued the matter. The story received good press coverage in several papers. In an interview the PR Officer at Vinmonopolet promised to look into the matter again, but commented at the same time that 'We would rather co-operate with the interest organisations in such cases’. This case illustrates a general problem that the visually impaired often experience, the problem of delegation. One comes as an individual to get information. If one is so lucky as to get a positive response in the form of promises to make information available, it often turns out that the organisation or firm concerned would rather act together with another organisation, as in this instance, the Norwegian Association of the Blind. After this incident had been widely reported in the press, Vinmonopolet finally made contact with e-com. They were not very happy about all that had been written in the papers. After a while e-com at last received the electronic text. It took e-com three to four hours to adapt the information. It turned out that what in the beginning had been claimed to involve enormous technical problems, was in reality simple to solve. The composition of the target group constituted, however, a central challenge. E-com’s target group was people with visual impairment, and this included both blind and weakly sighted people. In the project it was discovered that the needs and wishes varied greatly within the tar-

131

132

Researching ICTs in Context

get group. This was due partly to the degree of residual vision. Those with weak vision will more easily be able to relate to visual presentations, while information for the blind requires a greater degree of adaptation. There were therefore different views among the poorly sighted and blind about what was a good user interface. Many weakly sighted people very much wanted to keep up with technological development in the way in which it appeared to those who could see. Some visually impaired users were therefore provoked by e-com’s focusing on adaptation and made statements like 'I don’t want to be told what is best for me by people who see worse than me’. Correspondingly many blind and very poorly sighted people found it provocative that they were to adapt to a graphic user interface that seemed more of a hindrance than a liberator with respect to the possibilities of communicating. It is often claimed that when a particular technological solution fails, it is because it does not meet the real needs of the users. This type of statement illustrates a superficial understanding of the concept of need. When it comes to new technology one must very often see the solutions before one can formulate the needs. The concept of need in our context is thus relative and not absolute. Furthermore we can distinguish between two different contexts for the use of the concept of need, need as (1) a logical necessity in the sense that in order to talk to a person in another town in real time I have a need for a telephone, and (2) experienced needs or wants. Need is often treated as if it were stable over time, although few phenomena undergo so many changes over time and between different actors. Experienced needs may, incidentally, also be 'substitutes’ for other types of need. In most social experiments users are recruited according to a notion of the future user. However the problems of user studies include situations like the following (Woolgar 1994: 202): ! ! ! ! ! ! !

the user does not know his/her requirements the user knows his/her requirements but cannot articulate them the user changes his/her mind individual users say different things to different people users disagree as to what their joint requirements are individual users are not representative of (all) relevant users the user turns out to be a customer rather than just a user.

Experiments along the bazaar-route

Many needs studies in new technology are of little value precisely because there is no clarification of how different types of need are perceived and of the degree to which the actors represent different relevant social groups. One mistake that is often made is that heterogeneous social groups are categorised and common needs and aims are ascribed to them. The demonstration experiment of Internet solutions for the visually impaired illustrates how apparently homogeneous groups may interpret technology, needs and goals extremely differently. E-com resulted in a thorough debate within the user group, which again revealed different perceptions of how the user is included, or ought to be included, in the development of new media technology. The visually impaired therefore grouped themselves into different formal and informal groupings, who signalled different policy standpoints on their interests, but also different cultural understandings of new technology. In the next section I shall describe some standpoints that were established around e-com’s attempts to make adaptations to facilitate the use of new media technology. Even though I shall not be placing great weight on the emotional component of the three standpoints, it must be stressed that this component was at times extremely important, with 'flaming’ as one of several elements in the debate.

Framing and overflowing E-com was looked upon by those who took the initiative as an important democratic project. The democratic project was, however, contested by other actors. When e-com protested against the fact that others were 'stealing’ the documents they had adapted, the 'thieves’ protested, claiming that the information should be free for all. While e-com turned to the traditional mass media and mobilised traditional supporting players, e-com became the subject of debate and expressions of opinion in a number of contexts, not least on different computer bulletin boards. We therefore witnessed a marked overflow between frames and media. It was simply matters of chance that led to e-com’s being informed of this type of overflow. E-com claimed, in this connection, that it would have been natural for the debaters and the person with editorial responsibility to have given e-com the opportunity to answer various attacks. The feedback that was given was that e-com ought to know that people were dis-

133

134

Researching ICTs in Context

cussing e-com in various places on the net. The implication was that it was e-com’s responsibility to keep up to date with where the debate was being conducted. These debates demonstrate how difficult it is to make oneself the spokesman of a particular understanding of access and literacy. This understanding will quickly be problematised because one form of access is easily felt to be restrictive for others. Three frames dominated the debate.

'The Internet is a juggernaut that cannot be stopped’ It is difficult to resist development, so we must therefore see what comes and adapt the solutions of those who can see to those who are visually impaired. Adapting to supply often means that the solution to the problem is localised with the visually impaired person. The communication situation is to a great degree defined by the sender. In the course of the project period there was thus a large group who expressed views in line with the following: 'I don’t believe one can count on converting all the Windows programmes to DOS. That would be like believing in Father Christmas. You can’t stop development’. Adapting to supply was therefore perceived as important. This frame was linked to the metaphor of the information superhighway. Development was seen as something inevitable, to which one had to adapt in the best possible way, since one was not in any case among those user groups who would be able to set the agenda for how technological development was to be steered.

'The Internet is freedom’ Technology is both a tool and a life-style. This perception emphasises technology as a liberating force. One sees the user as an active force in the development of technology. The individual has far more opportunities than limitations. Experimental activity requires a wide framework, so one does not want any restriction on the freedom to experiment. As part of this notion great weight is placed on the freedom of the individual to try out and design new solutions. For certain relevant social groups this is in fact the core of the Internet technology or the Internet culture. In the project it was therefore clearly established that one would not be limited to 'e-mail’-like communication. This frame was linked to the metaphor of Cyberspace. One did not want to be told what freedom was, freedom was also freedom to define what freedom should entail.

Experiments along the bazaar-route

'The Internet is liberation from the body’ The initiative-takers behind e-com framed the Internet as liberation from the body. This frame was linked to the Cyberspace metaphor. The technological development 'was locking them out of the net’. It was therefore a superordinate aim to ensure that the body with all its weaknesses was made 'invisible’ in the technological development. Or as one of the initiativetakers behind e-com said in a newspaper interview: 'Most of my net friends don’t even know I’m blind’. Technology is therefore not interesting in itself, but it is important to focus on the social and societal role of technology. In this connection the dialogue between user and producer is of particular significance. Those who initiated the project had therefore problematised the development of graphic interfaces in relation to the visually impaired and had the aim of becoming spokespersons for the development of solutions for the visually impaired on the latter’s own premises. The three principles: 1) the pro-innovation principle, 2) the domestication principle and 3) the anti-diffusion principle, are central in all science and technology communication (Hetland 2001). The pro-innovation principle implies that an innovation ought to be diffused and adopted by all the members of a social system. Often this new benefit ought to be diffused more rapidly than is already the case. In contrast to the pro-innovation principle we find the anti-diffusion principle. This principle takes its point of departure in the fact that there is an innovation (or invention), but says that for different reasons this innovation (or invention) ought not to be either diffused or taken into use by particular user groups or by society in general. The principle in between is the domestication principle. This principle is a variant of the pro-innovation principle. That is to say that one does not reject the innovation, but takes as a starting point the idea that new technology entails great and important problems that must be solved before the media technology is taken into use in full. The disagreement about a particular framing can be explained in several ways. A likely explanation is that this notion reflects what one believes to be possible. The visually impaired have experienced the fact that media technology develops on the premises of those who can see. It can therefore be difficult to win support for the needs of the visually impaired. Adaptation to a certain degree of technological determinism may therefore be a natural option in such a context. However, also the more 'liberating’ notions have deterministic undertones. The first two

135

Researching ICTs in Context

frames entailed acceptance of the pro-innovation principle. The two groups that were most critical of e-com put the focus on the enrolment of new artefacts/properties; in their opinion the result should be more possibilities of choice for everybody. E-com focused on the domestication principle; both the technology and the users were to be domesticated to ensure a desired technological development. Carried out in a creative manner demonstration experiments may therefore have an important agenda function and may be a catalyst to bring forth different problematisations and thereby overflow between frames. Thus demonstration experiments show the variation in interpretative flexibility. During the past 20 years a number of social experiments with new technology have been set up. These experiments in many ways form 'hybrid’ communities in which the intention is to investigate how technology can constantly give the acting human being new possibilities to transform and develop the social and economic area. Linked to the 'diffusion of innovation’ model within traditional innovation theory, we find four types of experiment: (1) explorative experiments (2) pilot experiments (3) demonstration experiments and finally (4) replication or dissemination experiments. These four experiments are placed along the S-shaped diffusion curve. Technology becomes indispensable Replication experiments

Increase of user adoption

136

Demonstration experiments

Pilot experiments

Explorative experiments

Decreasing uncertainty about the innovation over time

Figure 1: Different stages in the experimental process (Hetland 1996a: 16).

Experiments along the bazaar-route

According to the diffusion of innovation model 'laboratory’ experiments are experiments at their most explorative stage, at which we can study the social relations that media technology is included in, or at least the relevance of media technology to different user groups. The aim of pilot experiments is to increase the attention focused on new technological possibilities, to stimulate debate and to open the way for the shaping of technology policy. Demonstration experiments are important tools in publicising and spreading media technology. They have often been used to promote special solutions among selected user groups to increase the general level of knowledge and thus also promote a more rapid diffusion of new technology. When technology is well known among both technology policy-makers and many user groups, replication or dissemination experiments have been used to disseminate methods that have been tried out, techniques or models, and they thus give the innovators local experience before full-scale introduction of the technology among new users. Different user groups participate in the social experiments and try to shape the results. In addition to negotiating one’s way to desired results and interpreting these in relation to specific interests, one also interprets and negotiates about the 'internal logic’ in the experiments. Seen from the participants’ point of view a 'laboratory’ experiment, pilot experiment, demonstration experiment or dissemination experiment does not imply locked roles. One can also negotiate about the role the experiment has been given as a starting point. There is therefore a clear tendency to translate the experiments in the earliest phases of the diffusion of innovation model in such a way that pressure arises to transform the experiments into later stages of the diffusion of innovation model. This is because some groups of participants are first and foremost interested in these experiments resulting in successful and lasting activities. Ecom was a demonstration project aimed to show that particular user groups needed particular user interfaces and particular forms of adaptation. To achieve the predicted success of the demonstration, the project was guided by a set of principles. First they offered low risks for the participants. In e-com this was attempted by strategies for the enrolment of all important national actors. Furthermore, they thought they had a rather good knowledge of adaptation, since they could build on the experience of several earlier failures. Secondly, the project designers tried to make the benefits of the new information system visible to

137

138

Researching ICTs in Context

the potential users. However at this point they had problems with the users since the users turned out to be a heterogeneous group. As a heterogeneous group they could not reach agreement on needs in common. The fact that e-com was in addition working to develop into a centre for electronic communication for the visually impaired meant that the door was opened for a conflict between different user groups. What therefore at the outset was defined as a specific type of experiment took on board problems that rendered the original definition of the experiment less meaningful.

The bazaar-route to technology policy The bazaar-route opens the way for hot situations. In hot situations most things are the subject of controversy, and these controversies are an expression of the fact that one does not have a stable basis of common knowledge and insight that one can agree on. In other words, what were at the outset good intentions invoked extremely different markings of interest. E-com failed in its attempt to establish a centre for electronic communication for the visually impaired. There is much to suggest that in this connection a perception arose that e-com was in the process of gaining a monopoly of communication/adaptation of information for the visually impaired over the Internet. Beyond the fact that there was disagreement about what adaptation actually meant, this perception suggests that other actors too were beginning to realise that e-com’s choice of media technology, i.e. the Internet, was an important strategic choice. The hot situations were useful in a number of ways. (1) Policy authorities gained insight into how heterogeneous seemingly homogeneous groups were, (2) The importance of the Internet was demonstrated to a host of user groups, who could again explore the net in their own ways, (3) The importance of adapting became central in public information policy; adaptation was nevertheless defined in relation to a far greater span of user needs than was the point of departure for this demonstration experiment. However, the demonstration experiment described has wider implications than simply creating better electronic solutions for blind and weak-sighted users. The experiment also illustrates the importance of facilitating interaction between different stakeholders in the process of developing new technology. Too often different stakeholders, latent or

Experiments along the bazaar-route

explicit, do not have their say in development processes with important consequences. In technology policy making it is therefore important to create arenas where stakeholders both learn how to express their needs and have their say. In other words, the bazaar-route is important in creating more hot technology policy making.

References Akrich, M. & Latour, B. 1992. 'A summary of a convenient vocabulary for the semiotics of human and nonhuman assemblies'. In Bijker, W. E. & Law, J. (eds) Shaping Technology/Building Society. Cambridge, MA: The MIT Press. 259-264 Callon, M. 1998. 'An essay on framing and overflowing: economic externalities revisited by sociology', in Callon, M. (ed.) The Laws of the Markets. Oxford: Blackwell. 244-269. Callon, M. 1987. 'Society in the making: the study of technology as a tool for sociological analysis.' In Bijker, W., Hughes, P.H & Pinch, T. (eds) The Social Construction of Technological Systems: New Directions in the Sociology and History of Technology. Cambridge: The MIT Press. 83-103. Cook, T. D. & D. T. Campbell 1979. Quasi-Experimentation. Design & Analysis Issues for Field Settings. Boston: Houghton Mifflin Company. Hetland, P. Forthcoming, 2002. Unmasking the Net: when technology communication turns to the public. Gøteborg: Nordicom. Hetland, P. 2001. 'The three faces of the Net: Internet narratives'. Paper presented at Computers at the Crossroads: Information society and beyond. Trondheim: NTNU. Hetland, Per 1996a. Exploring Hybrid Communities: Telecommunications on Trial. Oslo: Department of Media and Communication. Report no. 29. (Ph.D. dissertation.) Hetland, P. 1996b. Internett som inngangsport til offentlig informasjon. Elverum: Hedmark College. Report no. 5. Hofmann, J. 1995. 'Writers, texts and writing acts – constructed realities in word processing software.' The Mutual Shaping of Gender and Technology, University of Twente: conference paper. Tyner, K. 1998. Literacy in a Digital World. Mahwah NJ: Lawrence Erlbaum. Woolgar, S. 1991. 'Configuring the user: the case of usability trials.' In Law, J. (ed.) A Sociology of Monsters. Essays in Power, Technology and Domination. London: Routledge. 57-99.

139

140

Researching ICTs in Context

Woolgar, S. 1994. 'Rethinking requirements analysis: some implications of recent research into producer-consumer relationships in IT development.' In Jirotka, M. & J. Goguen, A. (eds) Requirements Engineering: Social and Technical Issues. London: Academic Press, Harcourt Brace & Company. 201-216.

Notes 1

This research was partly funded by a grant from the Norwegian Research Council under the programme 'Societal and cultural presuppositions for information and communication technology' and partly by a grant from Morgenlandet AS under the 'Programme for communication: Education, research and innovation'. Thanks to Helge Godø and Svein Bergum for helpful comments. 2 Cook and Campbell use the notion of quasi-experiment for experimental situations in which the experimenter cannot manipulate human behaviour. See Cook and Campbell (1979).

7 ICTs in contexts: reporting research from researching researchers' reports Beate Elvebakk

Orality and literacy In Walter Ong's classic study Orality and Literacy, he shows how writing is not just a tool but a way of life. I suppose the primary aim of the book is to describe the nature of oral societies, but I have always seen it as the story of how we – in our modern literal societies - have become what we are. I have continued to find it strangely compelling, probably because I have always been – like many academics, I imagine – a literary person, in the sense that I have a very close relationships with written texts. After all, we spend most of our working lives consuming or constructing texts. I think my fascination with Ong's book has to do with the way it describes an aspect of me that is at once incredibly intimate – the solitude of experiencing a novel you love, or a theory you can't get your head around – and strangely public – books being, after all, accessible to most people in our society. The interesting thing about Ong is that he is not content to point to our literal practices; he describes the literal mindset, and shows how our being accustomed to text to some degree explains what kind of people we are1. He does so by showing how incredibly different we are from people who do not relate to texts. In a society where information can only be stored in the heads of people, and only be exchanged in concrete interactional situations, information is simply something entirely different. It is so fragile, can so easily be lost or distorted, that you simply cannot take it for granted. It is a mode of action rather than a description of

141

142

Researching ICTs in Context

thought. Those who know – usually your elders – must be respected and listened to, whereas challenging old assumptions is a highly disruptive and dangerous act. One can't afford to take any chances with such a transient thing as knowledge (as information will always be in an oral society). We, having as a society had access to written sources for centuries, may be very different people. If we look for the truth about something, our first impulse is probably to look it up. We know that the truth about things such as the height of Mount Everest and the history of the Norwegian royal family can be found in books, preserved, ready to be picked up at any moment in time. We need not feel responsible for the preservation of knowledge – knowledge takes care of itself, outside our heads, in its own realm of stored and lasting information. The truth is out there. Literate societies also have a much more individualistic and decontextualised approach to knowledge than is possible in an oral society. Spoken utterance is addressed by a real, living person to another real, living person or real, living persons, at a specific time, in a real setting which includes always much more than mere words. Spoken words are always modifications of a total situation which is more than verbal. They never occur alone, in a context simply of words. Yet words are alone in a text. Moreover, in composing a text, in 'writing' something, the one producing the written utterance is also alone ( Ong 1982: 132). Thus literacy promotes solitude and introspection, and the fact that information has become virtually context-free may be seen to favour abstraction in thought. Because we can trust information to be maintained without our active participation in the process, we may also feel much more free in our relation with information. Information may be doubted, criticised, overturned or completely rejected. We know we can always find it, anyway. A literal society, having such a material and well-defined past, tends to emphasise the importance of progress. According to Ong, then, very much of what is most typical of our societies, and perhaps also most typical of us as individuals, has to do with the technology we have used for storing and transferring information. Due to the fact that information is so closely linked to the lives of humans, this change in information technology also spurned a shift in societal and mental structures. 'Technologies', Ong says, 'are not mere exterior aides but also interior transformations of consciousness, and never more than when they affect the word.' (Ibid.: 82).

ICTs in contexts

It doesn't take much imagination to see that similar points could perhaps be made about the introduction of ICTs. Like writing, and later printing, computer and network technologies enable a new paradigm of information storage, processing and transfer. And like writing, the new information technologies seem to embody a logic on their own.

Finding one's own place – or trying to I am currently working on my Ph. D. thesis, which concerns 'Researchers use of ICTs'. The researchers I have focused on are chemists working in the university of Oslo, and in order to map their use of ICTs as well as understand how these practices matter to their working day, their practical organisation of their lives, and their conceptions of various aspects of their own work, I have carried out fairly extensive interviews with about 30 % of the faculty. I have tried to focus what you might call the 'non-trivial effects of trivial uses of information technologies'; I have been trying to see whether the practices that are presented as very task-specific, goal-oriented and transparent, can nevertheless have unintended and important consequences for people's lives as well as for their practical working days. In short, I try to tease out an interesting interpretation from something that may be seen as uninteresting practices with ICTs and uninteresting views about ICTs. My informants are much more likely to send brief informative e-mails than experiment with identities, and access article databases rather than fanciful MUDs. They are much more likely to describe their use of ICTs as being a rational response to a communication need, than as a psychological experiment. Nevertheless, I would like what I say about them to be somehow more than ‘chemists send approximately 10 e-mails a day and receive about 20'. When writing a thesis, you generally want to produce something that goes beyond merely reporting what you observe, you want to say something that is truly interesting, to capture phenomena that have been hidden from view. This, of course, is very ambitious, but I believe these are the motivations of most researchers; whether or not we succeed is a different matter altogether. What I would really have wanted to do in my thesis is something similar to Ong's analysis, I wanted to produce the kind of description that makes us realise what changes our information technologies cause for us and in us, though in the rather limited context of a department of chemistry. However, I realised that this is not as easy as it may seem.

143

144

Researching ICTs in Context

In writing his analysis, Ong did have a few advantages that we do not: his analysis is a double comparative study. He could compare existing literal cultures to existing oral ones (admittedly a limited number today, but still). He could also – as his hypothesis is that these changes become manifest over time – compare early texts from literal cultures with texts from cultures that had used writing extensively for a long time. This opened up quite a big field of literature for inspection, and also enabled him and other scholars in the field to compare older preliteral mindsets with contemporary ones, thus reducing the dangers of confusing «effects» of literacy with spurious historical development. Ong's method, of course, is certainly not without its sources of error, but it does still possess massive advantages over our current situation. We do of course have access to pre-computer cultures, but our problem is rather the opposite – perhaps we do not really have access to a truly computer-literate society? There seems to be a relatively broad agreement that a new informational logic takes some time to develop, and that people's mindsets are not transformed overnight. The texts that Ong uses as evidence of an oral mindset, were usually produced – or at least written down – at a time when writing as a technology had been known for decades, maybe even centuries. So we cannot be certain that the changes that will later be associated with ICTs have materialised yet. March (1987) suggests that the most important effects of information technology will probably be unintended secondary effects, whereas we tend to overestimate the importance of intended and temporary effects. How can we tell then, what effects are significant and lasting, and which ones are temporary? Even if the important effects of computer literacy might have materialised, they would probably be extraordinary difficult to find. When studying the broader cultural implications of the use of ICTs, even in a very small community, numerical data is not in itself sufficient; you need to track people's responses and reactions. It is impossible to diagnose a cultural change on the basis of quantitative data alone. That’s why I chose to use interviews as my empirical input in my research on 'The use of Internet among university researchers'. I asked a number of chemists in the University of Oslo about their use of ICTs, and they were most helpful in answering my questions. I was told about how they use computers to run experiments as part of contemporary research methods in chemistry, how they process their data with computers, how they simulate experiments on their computers

ICTs in contexts

and how they carry out complex calculations in a matter of seconds. And I got to know about how they transfer data over the Net, how they access scientific information online, how they find literature, how they receive information, and how they correspond with friends and colleagues. ICTs are widely used, and I was able to get to know very much about its importance for the life of a modern-day chemist. I also learned about their fondness for email, their 'addictions' to the Net, and their frustrations about the technology. I was told about feelings of empowerment and about stress. I was not, however, told about looking at porn online, and only the young and inexperienced told me about playing computer games and reading newspapers online. One disadvantage of using interviews as your empirical basis is that they are likely to contain conscious distortions of the truth – people do not want to look bad. However, I did not have any reasons to believe that these distortions were so serious as to undermine the project, so I assumed that my general picture of chemists' use of ICTs was basically correct. Still, I felt that I was not much closer to finding the effects of computer literacy. The shift from paper to computers has been extremely swift, even in the natural sciences, where researchers have used ICTs longer than in any other sector of society. But even this very swift transition was very far from occurring over night; it was not something that took place at a certain point in time, once and for all. Basically, everybody is part of the transition – you cannot really distinguish between a «before» and an «after» generation. Everybody has one foot in each camp, so you cannot compare two distinct groups and their relations to information. Therefore, the only way to gain information about this transition is through asking people about their informational behaviour. What did they do earlier, what do they do now, how would they compare this to that, what are the major changes? These may seem to be simple questions, but the reverse is the case. Questions like these are incredibly hard to answer accurately and truthfully, as the use of ICTs is for the most part a «trivial» everyday routine, and not experiences that the users have reflected much on. How often do you check your email? How many emails do you receive daily? How many phone calls did you get in the seventies? How do you go about writing an article? These are things you do, but you mostly do them as part of other tasks, and these tasks are what you focus on, not the simple practices that go into them, perhaps constitute them. You

145

146

Researching ICTs in Context

give them a sort of secondary attention. They are the means through which you focus on the things you focus on (as when you use the computer for writing a letter. You don't focus on every single movement of your fingers – unless you are a very inexperienced typist – you focus on the sentences you are constructing.) The finger movements have become internalised. This is one of the reasons why we can expect that people are being changed by the technologies they use: they know them so intimately that they have become part of them. Michael Polanyi believed that knowledge was not the purely intellectual phenomenon it had been regarded as. He claimed that human existence in the world would always somehow be based on much more than our intellectual cognitions. Rather than saying that knowledge consisted in being able to articulate what we know, he claimed it was the other way around – that the things we know most intimately are very often those we find it most difficult to express. This has to do with the fact that the knowledge in question has been internalised, it has literally become part of us as physical beings. We know with our bodies, not our minds. Some parts of this bodily knowledge we may perhaps become able to articulate, but it is impossible to exhaustively describe this realm of knowledge. To know then, consists to a great extent in being able to do, and we do the things that have meaning for us. Thus, our knowledge will always have its foundations in our lives. Knowledge is not an external system for describing the world, but a fully internalised way of coping with it. When humans know something, this cannot reasonably be seen as a representation of the world. This means that much of my informants' knowledge of their own use of ICTs is more or less 'tacit', in the sense that they rarely articulate their practices, and when practices aren't articulated, they are not known on the articulate level. Even a very concrete question like how many emails one receives or sends every day is often modified within the course of an interview, because some are forgotten because they are irrelevant, some because they are dealt with immediately, and some because they are very brief messages. In my interviews, I would typically ask at a relatively early stage how many e-mails the person received a day, but this number would often be modified when the person went on to list the kinds of -emails received, as it became obvious that the number initially reported was to small. The same goes for use of ICTs in the past; what has not been articulated is extremely difficult to remember. How we use the different technologies

ICTs in contexts

that we use as part of our job is not something we think about, unless it is for some reason surprising, or inconvenient. As long as everything works, we do not normally spend much time considering how or why it works. This feature of our relationship with the things that surround us makes it very hard to say anything about development over time, as long as we use interviews as our input. When people are not able to describe how they use information technologies today, let alone in the past, it is very hard to compare the two to see what has happened. And still we're just dealing with the practical, measurable aspects if behaviour vis á vis ICTs. Most of the changes described by Ong are not practical; they have to do with structures of actions and thoughts that are related to information, but they are not the same as the actions and thoughts that immediately concern information technologies. If we have turned into the kind of people for whom the durability of information is taken for granted, this is not obvious to an outside observer from what we do. It may, of course, become apparent in the things we say, as when we talk about looking something up in an encyclopaedia, or about rejecting the contents of a specific book. Nevertheless, we say incredibly many things that are to do with information, so how could an outside observer select the significant ones, from such an unsolicited mass of information? As a researcher looking for the cultural significance of ICTs, this is more and less your situation as well. You receive an enormous amount of information; hours of people's thoughts about their use of ICTs, but you cannot really tell what is important, and what is not, what points to a cultural development, and what is personal whim, what development has to do with the use of ICTs, and what is related to other trends. Do chemists communicate with foreign researchers more frequently because of the ease of electronic communication, or because their funding is tied to collaborative projects in the European Union? Do they read more articles because of the access to huge library catalogues online, or because their field has expanded? It can be incredibly difficult to establish causal connections even in a laboratory where you are dealing with a relatively closed system. It is virtually impossible in an open system like our society where changes are always paramount. So you may find that we already have the following problems: you cannot trust your informants to provide you with correct information, their knowledge about their own practices is limited, and often incorrect, and their memory of past practices is even less trustworthy. Even

147

148

Researching ICTs in Context

if your informants had been able to inform you about past and present behaviour, you cannot straightforwardly observe the changes we are talking about – they are to subtle to be directly measurable. (This might have been easier if you knew in advance what changes to look for - but you probably don't. Would someone from an oral society been able to predict the effects described by Ong?) You do not merely want to know how many emails someone receives every day, and how many letters they received in the past, you want to know if this changes something more fundamental about his or her relationship to the world. Further, even if you could ascertain past and present behaviour without a shadow of doubt, and you knew exactly what changes to look for, you would probably not be able to establish a causal link, as the introduction of ICTs has happened gradually over years or decades, and many things have changed in that period that might influence our relation to information. Anyway, the effects are likely to appear only after some time has passed, which makes them harder to link with the technology in question. Amazingly, things get worse. When you do interviews, you do not, as we have pointed out, have any direct access to the truth per se. You have access to one person's understanding of reality. This means not only that you have to trust this person's reports about his or her practical actions, with the conscious and unconscious distortions that they are likely to contain, but you also have to be aware of the fact that these reports are already interpretations. When someone tells you about their life – however trivial the details – they tell you something that has already been fitted into a story. What your informants tell you about their use of ICTs is thus already an interpretation of this aspect of their lives. This goes for the practical details of their work as well as for their feelings and reactions. The philosopher Charles Taylor has written extensively about human life as interpretation. In the essay 'Interpretation and the Sciences of Man', he tries to show that any social science, however quantitative the approach, is always interpretative. The reason for this is that the objects of study – human beings – have always already interpreted themselves. The actions they do always make sense to them in a certain way, however irrational, they have meaning, and they can only be made sense of if the researcher is aware of this. Humans are 'self-interpreting animals' that relate their lives to a field of meanings, and these meanings must be the starting point of the researcher, whose goal must be clarification of these existing interpretation. Thus any social science is a re-interpretation.

ICTs in contexts

His example is that of an election, which is apparently a quantitative process, where the actions are bound to a given and static interpretation: the researcher may count the votes cast for each of the candidates. Taylor, however points out that this is so only on the background of a specific interpretation of the situation: Thus, in voting for the motion I am also saving the honour of my party, or defending the value of free speech, of vindicating public morality, or saving civilisation from breakdown. It is in such terms that the agents talk about the motivation of much of their political action, and it is difficult to conceive a science of politics which doesn't come to grips with it.' (Taylor 1971: 19). If we want to approach these questions 'scientifically', through using a questionnaire to map 'values', we will have to base the questionnaire on our interpretation of the actors interpretations of their actions. It is simply impossible to provide meaningful explanations of people's acts without making an interpretation, and even a possibly flawed one. Description of human action is already interpretation. Thus, not only is every statement in your interviews already interpreted, but the brute facts you are told about are also in part an interpretation of a situation. The actions described are the responses to a given situation – as interpreted by the actor. When a professor tells me that he (it's usually a he) doesn't visit chemistry newsgroups because he doesn't have the time, this implies as certain conception of newsgroups (as 'surfing the net', wasting time), and it suggests a certain understanding of the role of the professor, and it also probably indicates what image he wants to project of himself to the researcher who asks all the questions. As a researcher, you only have access to this at all because your interpretations are often similar to those of your informants.

Past masters We seem to have encountered insurmountable obstacles to our ever finding the truth about how ICTs influence our lives and our understanding of ourselves. Even the reporting of practical behaviour involving ICTs is not without difficulty. Is it still possible to say anything positive at all about this process? Still, quite a number of theorists have tried to do just that. Writers like Manuel Castells,2 Sherry Turkle and Jay D. Bolter have all based

149

150

Researching ICTs in Context

entire careers on trying to make sense changes associated with computing and networking, and, to my mind, all three seem quite nonplussed by the difficulties sketched above. So how do they go about the task?3 Naturally, it is not possible to provide a description of their work that does the writers justice within the confines of this chapter, but I will try to give an rough sketch of their respective approaches. Let's start with Manuel Castells. He gives an enormously detailed and wide-ranging description of the emerging 'network society', of which network technology is only one of several enabling factors. He does not only describe factual changes in the political, economical and cultural sectors of societies, he simultaneously suggests new models and metaphors for understanding and making sense of the upheaval. Though the starting point of his work is strictly empirical, and he illustrates his points with numbers and statistics, his interpretations are also imaginative, as when he postulates the emergence of 'a space of flows' and 'timeless time'. Though making very active use of the extremely extensive factual basis for his work, he ties the different elements together in ways that cannot in themselves be found in the empirical material. In using the concept 'network society' as a synthetic force throughout his analysis, he constructs a framework for understanding a wide array of developments in a more unified manner. Sherry Turkle's approach is very different, although her book Life on the Screen, is also based on empirical work. Her focus is primarily psychological responses to computers, and how our use of computers influences our identity and our ideas about our identity. In her own words, her book 'describes how a nascent culture of simulation is affecting our ideas about mind, body, self and machine.' (Turkle 1996: 10). She uses interviews more as illustrations of her theoretical points than as absolute evidence of their correctness. When she describes the 'culture of simulation' she sees to be emerging, she builds on earlier theoretical writers such as Baudrillard and Lacan as well as on her personal experience with computers. When using interviews, the emphasis is on the interviewees' experience of their own situation, rather than on factual information. She also situates her informants, relating their stories – their computer-use-history as well as significant events and patterns in their personal lives. Although she stays close to her empirical material, its role is somewhat ambiguous, oscillating between being motivation and illustration for her theoretical framework.

ICTs in contexts

Bolter's work in Turings Man was published a few years before the other books discussed here, and thus written before some of the latest breakthroughs in digital technology took place (Endearingly, his introduction contains the sentence 'Most laymen have never been in the same room with a computer.'). Its approach is explicitly historical, drawing on the author's background as a classics scholar. He describes the computer as a 'defining technology' (an honour shared by the potter's wheel in the ancient world, the clock in medieval times and the steam engine for the industrial revolution), redefining the relationship 'of mankind to the world of nature. (Bolter 1984: 9). It also recreates the relationship of mankind towards itself, however, when we 'come to think and speak in terms suggested by the machine', such that eventually 'man recreates himself, defines himself as a machine'. In support of his thesis, he draws analogies to other historical epochs and provides a relatively thorough description of the working principles of computers. These approaches may at first glance seem very dissimilar, and thus perhaps not very helpful for furthering our understanding of how to do research about ICTs. There are however some common features that I believe we can learn from: !

To some degree, the writers all base their observations/predictions upon qualities of the technologies themselves. When Castells postulates the existence of a 'space of flows' or 'timeless time' he uses empirical examples for illustration, but the concepts owe more to reasoning than to systematic observation. Because the technologies open up these possibilities, and because financial and other considerations causes us to make use of them, this is what happens. Likewise, when Sherry Turkle sees modern computers as embodiments of post-modern theory, she does so by pointing to the computer's changing logic, the increasing reliance on surface, the 'triumph of tinkering', the fact that deep understanding has become redundant. Excerpts of interviews are used to enhance her thesis, but they do not carry the entire weight of the argument. Similarly, Bolter provides a description of the logic behind the computer, in order to show how this in turn influences the ideas of its users.

!

Neither author attempts to establish a strict causal link, rather, they try to draw a broad picture, in which their empirical find-

151

152

Researching ICTs in Context

ings fit in, but for which they are not in themselves sufficient evidence. They do not postulate absolute causality, and do not try to 'prove' this connection. Castells suggests that ICTs 'open' certain possibilities which are, for different reasons, often seized upon. Turkle never does specify how – through what exact mechanisms – computers affect identity. Nor does she discuss whether the culture of simulation is somehow a result of the existence of computers, or whether this nascent culture somehow influenced the development of the modern computer. Similarly, Bolter sees a new image of man emerging, but, as he points out, that defining technologies have not always been the most practically and economically significant of their respective ages – we cannot really explain why this technology rather than another serves to define humankind in our time. All he can say is that it must somehow 'appeal to the mind' of the age. !

The authors do not necessarily expel other working factors. Castells explicitly emphasises that the developments he describes are the outcome of several unrelated processes, of which the new technological opportunities opened up by the Internet is only one. When Turkle introduces post-modern theory, she implicitly accepts that the culture of simulation was well under way long before the introduction of the Internet. Bolter maintains his 'appeals to the mind' hypothesis, which does not answer what features of our age that render us receptive to this picture of mankind.

!

Each of the writers is aware of the interpretative situation. They realise that what they are working with are not mere 'brute facts' but also people's way of interpreting situations. The interpretations of the researchers enhances the subjects own interpretation of their actions, rather than rejecting them or accepting them uncritically.

!

These three scholars are actively aware of their own role as interpreters. Castells informs the readers about his feelings for some of the groups he writes about, Turkle actively engages her own experience with computers in the story, and Bolter emphasises how his background as a classics scholar influences his interpretations.

These common denominators may be perceived as weaknesses of these books, and if you are a strict scientist this is probably how you

ICTs in contexts

will understand them However, I believe it is no coincidence that they have been able to defend their position in a field that has otherwise moved very fast, so I conclude they must be strengths. They may have sacrificed some «scientific rigour», but they have gained the power to capture the imagination of their readers.

Early Errors However, there is certainly an abundance of examples to be found should we look for failures. Without mentioning names, I would like to point a finger at some of the earliest chroniclers of ICTs. I think many of them shared some common traits that have made their books and articles date very fast, so that they are virtually useless for a researcher reading them today. I can't claim to have constructed an authoritative and exhaustive list, but these are, in my experience some of the most typical weaknesses that we encounter time and again: !

Mistaking early users for typical users. Early literature on the Internet made a number of assertions about the 'effects of technology use' – often well supported by statistics and measurements – that did in reality probably say much more about the early users of any technology, than about any effects the Internet might have. Contributions within my field include claims that use of the Internet leads to higher publishing frequency, and that use of discussion groups will grow exponentially and replace traditional conferences.

!

The related mistaking of very intensive use for typical use. In their search for the effects of technology, many researchers turned to those who used it most extensively, such as hackers and other teenagers who spent most of their free time online. They often ignored that the effects they found could be related to the group in question, and that they could therefore not necessarily be seen as indicative of use by other groups in the future.

!

Researching MUDs for exotic topics and easy access to research data. This point speaks for itself. An unproportional amount of research on ICTs centres on MUDs (multi-user dungeons), which is still a relatively marginal application. Studying MUDs is tempting, as they provide exotic 'worlds' to which the

153

154

Researching ICTs in Context

researcher is given absolute access, but they are nevertheless only a very small part of the picture. !

Being misled by Sci-fi to seeing 'cyberspace' as separate realm. This is probably related to the exaggerated focus on MUDs. In a MUD, one may easily imagine the cyberspace as a different, fictional-yet-real dimension, that exists as a parallel to the real real one. There is also the fascination with technology being 'predicted' by literature. This way of understanding the Internet is however very rare amongst the majority that uses the Internet relatively sparingly, and mostly for well-defined practical tasks. This assumption has therefore made many early articles on the Net strangely irrelevant for describing the experiences of most users.

!

Predicting changes on the basis of enabling technology alone. We have seen that centring once analyses around the features of the technology in question can be fruitful. But it may also be dangerous, if one does not carefully think through one's ideas. Many predictions about Internet use seems to assume that existing structures are upheld by obvious practical restrictions only, so that an enabling technology will necessarily lead to restructuring. In this vein, a number of different writers believed that the academic journal of the 21 century would be liberated from the reins of greedy publishing houses, and finally be freely available to anyone interested. Similarly, quite a few writers have also foretold the demise of the university, as information is now equally accessible to all, wherever they are.

!

Unbridled enthusiasm. Much literature about ICTs has been overly optimistic about the possibilities opened up, rather than trying to get the balanced view through careful study of actual use in contexts.

!

Luddism. The opposite stand. Sees ICTs as having dire consequences for society and individuals, without looking into actual use.

If we are to sum up, we might say that many of these faults are to do with the uncritical use of empirical inputs, so that it is either given to much weight, without being placed in a broader context and critically examined, or ignored altogether. Today, our heroes are those who have struck the balance right.

ICTs in contexts

Many of these studies probably exhibit much more scientific rigour than the books we examined in the previous section. The reason for this is that they have often chosen to look at a very limited number of elements. If you examine whether there is a systematic relation between e-mail use and publishing activity, for instance, you have reduced the number of variables to two, and even if you may check for other possible reasons for this covariation, you are still dealing with a very limited number of elements. You may almost be said to have constructed a closed system ,as so most factors have effectively been excluded before empirical work commences. As a matter of fact, most of the errors listed above betray the same tendency: the researchers have chosen to limit the factors they look into very severely, rather than opening up for a wide view; they focus narrowly, and thus in some sense get a closer view. I believe, however, that the view may be too close, so that one loses one's perspective.

Striking a balance when studying ICTs We have seen that making constructive use of empirical findings can be very difficult. Yet this, presumably, goes for anyone working in the social sciences. Is the predicament different and somehow more pressing for those of us who try to say something interesting about information technologies? In a way, I believe it is. For one thing, as we have discussed, our object of study very often concerns practical aspects of people's lives, which makes it harder to obtain reliable reports. In some cases one might choose to opt for participant observation, but in many instances, the use of ICTs is a solitary business, where an observer would be experienced as very obtrusive. Even if one uses observation rather than interviews, some of the information one is looking for will not be observable, and it will require another approach as a supplement, where the same problems will resurface. A second point has to do with the rhythm of development – in one sense, we have a 'computer revolution', in a another, this seems to take a lot of time, as things are continuously undergoing change. We are aiming at a moving target, but one that moves sufficiently slowly for most people to experience it as a gradual process, and thus not stop to consider radical changes4. So we are not dealing with a given and static situation, and at the same time we are dealing with processes that are slow enough to remain mainly unnoticed and untheorized by those involved5 In addition, as we have seen, we may end up describing a 'phase' rather than something essential.

155

156

Researching ICTs in Context

Thirdly, writing about ICTs often – whether intended or not – hurls the researcher into some kind of futurology. ICTs will probably, in one form or another, be used extensively for a very long time to come, and thus any hypothesis about their impact not only has to hold its ground under presents circumstances, but is likely to be confronted with a slightly altered reality in the future for which, ideally speaking, it should also be of relevance. (Anyone working in this field has come across the embarrassingly optimistic writings of the early to midnineties, where the researchers habitually claimed that peripheries would become as central as centres, and that authority would automatically evaporate, and flat structures become the rule. Ideally, one would like to avoid such blunders.) To make matters worse, ICTs have somehow for decades maintained an image as «the technology of the future», and this often leads people to assume that anyone researching ICTs is in the business of fortune-telling. Finally, so far, the study of ICTs has been truly interdisciplinary. Researchers have come from a multitude of fields, including psychology, sociology, literature and linguistics. There is no established canon to which the inexperienced researcher may turn to find the correct method. Taken together, I believe these points makes the situation somewhat more pressing for the student of ICTs than for your average social scientist, and therefore we are probably in a more precarious methodological situation than most other disciplines. We may accept this situation, or we may not, but the more interesting question is how we rectify these shortcomings. Some answers are relatively straightforward. If the problem is the lack of correct factual information about patterns of use of ICTs, we may collect correct information through the technological logging mechanisms. (Though the availability is somewhat reduced by strict Norwegian privacy regulations). This kind of information can serve to complement information received in interviews, but can certainly not replace it, as it does not indicate context, such as reasons or motivations for factual use. As regards comparing with past practices, in order to suggest causality, there is often some relevant historical information available. Our field is quite new, but other information practices have often been studied in the past. These do not concern the exact same people, and often not the exact same group, but can nevertheless serve as an indication, if used with caution.

ICTs in contexts

The more serious problems remain. We cannot directly observe the phenomena we are looking for; they are not there to be picked up by us. Causality remains tricky; it is incredibly hard to establish that anything at all is 'caused' by digital technology. And we cannot get beyond people's interpretations of their own lives and their own actions, let alone our own. So how do we go about our tasks as researchers once these considerations have been accepted? These are my suggestions: !

Construct frameworks. I believe we have to accept that our position as researchers requires us to construct frameworks, and frameworks that owe as much to our own interpretation of the situation as it does to the brute facts we are studying. We should, of course, try to invent frameworks that 'fit the facts', but we cannot expect to find them in the facts. We derive our frameworks from the empirical facts, but the frameworks are not caused by them. If we do not go beyond the facts, what we are saying is probably not very interesting.

!

Accept the tenacity of 'facts'. We are already part of an interpretative tradition, and as such we are also equipped to expand on others' understandings – but not in a completely objective manner. When using empirical data, we should take care that we neither under- or overestimate their importance. They need to be seen in context, so that we constantly draw on what alternative sources there might be, but we must also realise that we are never going to hit 'rock-bottom' – we always relate to materials that are already interpreted, by others or ourselves. This means we should also try to be reflective and reflexive concerning our own position as interpreters.

!

Place ICTs in their proper context. Do not overlook the structures in which they are used, and beware of the motivations and constrains deriving from offline phenomena and events.

!

Use properties of technologies as a starting point, but don't stop there. Complement with critical analysis and empirical inputs. (Do ICTs undermine the importance of space? Find out with whom we do communicate, what space means beyond distance, etc.)

!

Realise that developments are usually complex, so that it is quite unlikely that you will find a neat, linear story.

157

158

Researching ICTs in Context

These are probably not revolutionary ideas, and may seem rather obvious. I believe, however, that the quality and level of reflection in the average article about 'the social aspects of computing' have been raised quite significantly since the early days of the Internet, when this field may have been considered an easy route to funding. They should nevertheless be kept in mind.

The tricky question of truth So, we have established that, at best, we mostly do have access to people's edited interpretations of how their reality has changed. And our work as researchers is to make a theoretically interesting interpretation of this interpretation. We have reviewed some instances where the researchers have succeeded in so doing, and we have tried to capture the secret of their successes. We have seen that those who construct interesting theories about the influence of ICTs in our lives do not seem to use the empirical as their absolute starting point, on the contrary, their hypotheses go beyond their empirical findings, and are more ambitious than the material allows for, strictly speaking, and we have concluded that this is probably what must be done if one wants to say something interesting about ICTs. However pleased we might be with the resulting analyses, this still leaves a tricky question: how much 'truth' is there still left in our final work? According to the points listed above, facts are not mere facts, and frameworks should move beyond their factual bases. This seems to suggest that the resulting analyses will be less 'true' than we would ideally wish. Our work will not be direct reflections of a pre-existing reality, but to a certain extent constructs. This leads us to a further question: should the truth be a goal for our research? In one sense, I think we would all say it should. Research, after all, is about finding the truth, and if we are not after the truth – why bother with doing empirical work in the first place? On the other hand, I think Taylor's insights are extremely important. Our business is interpretation, and we cannot go beyond that. There are no absolute hard facts to be found. Searching for a truth beyond interpretations is probably futile, and certainly unnecessary – we manage well with the relative certainty that we can still obtain.

ICTs in contexts

This is why I think we should focus on some of the other points above, the ones dealing with contexts and using multiple sources. To see a technology in context means to include a number of different variables, ranging from the characteristics of the technology to the interpretations of the users, and the users' non-digital contexts. In this sense, I believe that in becoming less 'true', our work may become more 'truthful'. A narrow focus on decontextualised facts is more likely to eschew the truth than a reflexive, interpretative study that uses facts with some caution. There is also another – and more concrete - sense in which we may be said not to approach truth. As mentioned above, there tends to be an element of futurology in what we are doing; mere descriptions tend to be read as prediction. Thus, there is a fair chance that what we write today will serve as illustration of classical mistakes for tomorrow's students of ICTs. Perhaps we should accept that at this stage in our young 'discipline' we should be content if we can construct a relatively convincing hypothesis, a metaphor that enables people to make sense of this new thing that has happened to us. If the fate of this picture is eventually – or immediately, which is probably more likely – to be rejected or refuted by another researcher, we may still have contributed to someone else's insight.

References Bolter, J.D. 1984. Turing's Man. Western Culture in the Computer Age. Chapel Hill: The University of North Carolina Press. Castells, M. 1996. The Rise of the Network Society. London: Blackwell. Kraut, R., Patterson, M., Lundmark, V., Kiesler, S., Mukopadhyay, T. & Scherlis, W. 1998. 'Internet paradox. A social technology that reduces social involvement and psychological well-being?'. American Psychologist, No. 9, September. 1071–1031. Kiesler, S. & Sproull, L. 1987. Computing and Change on Campus. Cambridge: Cambridge University Press. March, J. 1987. 'Old colleges, new technology'. In Kiesler, S. & Sproull, L. (eds) Computing and Change on Campus. Cambridge: Cambridge University Press. 16-27. Ong, Walter. (1982). Orality and Literacy. The Technologising of the Word. Routledge: London.

159

160

Researching ICTs in Context

Taylor, C. (1985). 'Interpretation and the Sciences of Man'. In Philosophy & the Human Sciences: Philosophical Papers. Vol. 2. Cambridge University Press: Cambridge. 15-57. Turkle, S. (1996). Life on the Screen. Identity in the Age of the Internet. Phoenix: London. Zuboff, S. (1988). In the Age of the Smart Machine. New York: Basic Books.

Notes 1

Not being in a position to speak for Ong, I will speak only for myself. I would like to point out that I do not believe this position implies a determinist stance towards technology. The fact that the use of written texts seems to have much the same 'effects' in most literal societies is probably related to the fact that these societies share a common understanding of how the technology of writing should be used. In different societies, that used writing exclusively for religious or magical purposes, say, the 'effects' of writing would probably be very different. Similarly, societies where only a small elite is literate will probably not witness all the effects described above. 2 I am aware that Castells' analysis of the network society is meant to capture trends that are not necessarily caused by technology alone, but as he does to some extent attribute these developments to the use of network technologies, and as I believe the point about causality to be valid whether the causes are technological or financial, I will still use him for illustration. 3 These theorists are chosen relatively randomly, because they still are among the most oft-cited authors in this fast-moving trans-disciplinary field, even though their work is no longer entirely new. Relatively speaking, I think their books may be referred to as 'classics'. 4 There are some examples of studies of computer-use where the researchers have carried out a 'social experiment', and introduced ICTs to new users, thus having the opportunity to compare 'before' and 'after'. Among these are the early writings on the introduction of the Internet at Carnegie Mellon University (see for instance) and the more recent work which served as basis for claiming that the Internet reduces social and psychological well-being. (Kraut et al. 1998). 5 Zuboff (1986) sees her study as enabled by the transitional situation in which her informants find themselves. But her informants relate to a very abrupt change, as they are employees whose entire technological surroundings have been replaced overnight by ICT-based tools and processes. Such a transition leads to reflection around differences, something that does usually not take place to the same degree when changes are slow, gradual and wilfully chosen.

8 The multi-dimensional stories of the gendered users of ICT Hilde Corneliussen

Introduction Gender has been an increasingly important topic within research on ICT during the last two decades, at the start dominated by questions about what computer technology did to women, and then by questions about what women did not do with computers. In my Ph.D. project, on which this chapter is based, I am not primarily asking what men and women do or do not do, but what gender means to men and women with a relation to ICT. The working title of the project is Relations between people and information and communication technology (ICT): how does gender affect these relations? The main goal is to study how individuals construct their relations to computers, with a special focus on how gender makes a difference within this construction. The historian Joan Scott, who has inspired my gender perspective, claims that it is the historian's task to follow the endless struggle of meaning, and to explore how some meanings acquire the nature of truths. (Scott 1988: 5) I am not following this 'endless struggle of meaning' in historical documents as Scott proposes. Instead, I am searching for that meaning which the informants – students at the Department of Humanistic Informatics – perceive as valid; in other words, I am looking for the meaning of gender in relation to computers at a specific point in history.

161

162

Researching ICTs in Context

Questions raised by a preliminary study In a preliminary study at the Department of Humanistic Informatics in 1997 (Corneliussen 1997), I found a pattern in the way men and women presented their relations to computers; men expressed a confidence towards their own abilities to work with computers, while the women expressed a lack of confidence. This pattern did not correspond with observations I had in the computer lab. The mismatch between people's practical abilities and self-presentation raised the question of how women could ignore their own abilities while men so easily could talk about themselves as computer competent, even when they were not. Were they expressing some kind of personal inconsistencies? And why did this mismatch present itself in a gendered pattern? In this chapter I will discuss the challenges that these questions imply, that is the search for a way to understand what seems to be multidimensional stories of individuals' relations to the computer. It indicates that this study, which is positioned in the crossroad between gender research and technology research, has to take into consideration a complex structure of factors in order to understand how gender and technology are affecting each other. It illustrates the need not only to ask what people do with ICT, but also how they experience values and ideals connected to ICT; that is, how they experience what they do. This is the background for the key questions in this project: 'How is gender negotiated in relation to ICT?' and 'How is ICT negotiated in relation to gender?' I start with a short presentation of my empirical material, after which I present the theoretical and methodological basis for my gender analysis. In closing, the method will be exemplified through an extract of a reading of two informants' articulations of their relationship to computers.

Empirical material The empirical material of my Ph.D. project is based on a three-month study at the Department of Humanistic Informatics. It includes e-mail questionnaires and observations of 30 students, 8 men and 22 women.1 The selection of participants was based on voluntary participation by the students I met as a lecturer at the beginning of the spring term 2000. Each week for about 3 months I sent the participants a new e-mail questionnaire, with questions about previous and present experience with computing, and about the meaning of gender in relation to computing.

The multi-dimensional stories of the gendered users of ICT

The students belonged to three different programming classes that I was teaching at that time, and I used these classes to observe the participants in their work on the computer and in the groups as well. At the end of the term I interviewed 23 of them, 7 men and 16 women, in groups of 3 or 4, a selection based on the degree of participation in the other parts of my study. The interviews lasted between one and one and a half hours, and followed a set of predefined questions to ensure that I would touch on to the same topics in all groups. The interviews were later transcribed.

The challenge of sharing context with informants To use informants within a context where I have several positions involves a special challenge, and in this section I discuss my way of dealing with this challenge. I primarily concentrate on the question from a theoretical and methodological perspective, and I leave the discussion of the final results to my forthcoming thesis, as they are connected to the ongoing process of my work. The question of how scientific knowledge is produced has been a central theme within the philosophy of science since its inception, and the question of the researcher's position has also been a part of this long lasting discussion. The notion of the objective researcher as it has traditionally been found within the Natural Sciences, but also as an ideal within the Social Sciences and the Humanities (Kjørup 1996: 91), has been criticized from various positions, and especially by feminists (Asdal et al. 2001: 17). The philosopher Sandra Harding and the biologist and philosopher of science Donna Haraway have delivered some of the most important feminist contributions within this debate, both dealing with the theory of science and technology research. Although their contributions are not identical, they both emphasise science as a localized practice which needs to be contextualised (cf. Haraway 1997: 36).2 I will present this view by using Donna Haraway's terminology, a vocabulary which alongside her work has been been a source of inspiration for my research.

The invisibility of the modest witness 'Modest witness' is the name Donna Haraway has given the objective researcher, who is also an invisible researcher. This researcher is the 'modern, European, masculine' scientist, who could claim objectivity purely through his inscription in a traditional research institution, ac-

163

164

Researching ICTs in Context

cording to Haraway. He appeared as the 'witness' of 'the object world, adding nothing from his mere opinions, from his biasing embodiment.' (Haraway 1997: 23-24) Haraway challenges this scientist, or tries to 'confuse' him in her characteristic style of writing, in order to create a more 'corporeal, inflected, and optically dense' (Haraway 1997: 24) modest witness. Haraway's message is that the modest witness under no circumstances is invisible, and should therefore take responsibility for his own influence on the knowledge process. The modest witness can make his position within the research process become visible and known through reflexivity or situating his position and contextualising his vision, a process in which Haraway's concept of situated knowledge has been employed (Haraway 1991, cf. Berg 1998: 76-79. See also Rustad 1998). This discussion deals with my effort to become a situated researcher, to clarify my location and vision, a process described by Haraway as 'the always partial, always finite, always fraught play of foreground and background, text and context, that constitutes critical inquiry.' (Haraway 1997: 37).

Which positions do I activate in the context I share with my informants? A point I return to later is that we all hold a number of subject positions3 simultaneously, but these are activated in different situations. I need to consider which positions I activate in the situations that I share with the informants, because these positions might have affected the informants' articulations. The point is to detect any possible influence I might have had on them, and not to avoid or hide this influence. The most important positions I had within the group of informants were that of a lecturer in classes of 16-18 students, researcher, and woman. Do they answer the way they do because they see me as a researcher or a lecturer? Or because they see me as a woman? Or considering the most traditional notion about computing as masculine and male dominated, are the answers affected by me being a female lecturer in programming? In the analysis, I have to be aware of which of my roles they are addressing, and it is not every articulation that address my positions as clearly as these examples: 'I will start to work harder with programming next week,' clearly addressing the position of a lecturer. 'At the university, it is normal to talk about gender as a dualism…' which

The multi-dimensional stories of the gendered users of ICT

seems to be addressing a researcher, or in '…because you are a female lecturer', where I am being addressed as a woman. The way the informants answers my questions can also tell something about how they perceive the meaning I produce in my articulations. What do they do with the messages I send out? Do they accept the questions as 'valid' or interesting questions to ask? Do they answer the questions I ask or do they answer a question that I did not articulate? Do they accept and repeat my concepts, or do they find new words to cover the topics I introduce? All of the questions raised here are treated in the practical analysis of the material, of which my articulations are also a part, interwoven with the informants' articulations, and analysed together with them. I have also been working with these questions from the beginning of the project, for instance through my research notes from my classes and meetings with the students, during which I related my observations to my thoughts and expectations towards them as students. One of the things I can see in these notes is that despite my own awareness that gender as discourse does not have to be identical with 'practiced gender', some of my expectations clearly followed a traditional gendered pattern, for instance by being more impressed by clever female students than by male students and vice versa, to be disappointed by male students who did not follow my teaching as fast as I expected. In this way these notes became a medium with which to increase my awareness of which discourses about gender and computing were internalised as hegemonic in my own thought structure, and thereby as correctives to my readings of the empirical material. In addition, there are the discourses or messages that I try to spread among the students as a lecturer. However, after several years with these courses, I have a clear idea of which 'messages about computing' that I in general try to communicate to my students.4 I mentioned my position as a woman as one of the important positions I hold, and this is perhaps harder to define as a context outside the text; this is not a position that has materialised itself in text, and it is rarely addressed directly by the informants. But still it is both important and interesting to ask, not whether a man would have asked the same questions, but whether he would have received the same answers. The positions presented here are not the only positions I might have, but the messages conveyed by all of my positions have one thing in common: the informants' freedom to agree with them, reject them, alter them – or to

165

166

Researching ICTs in Context

ignore them altogether, a point that is illustrated within my material. Furthermore, it is too simplistic to believe that my positions alone produce the informants' answers or attitudes, or that an experience of 3 months duration has produced their discursive understanding of how gender matters in computing, as they all have their own previous experiences and understandings, or 'moral economy'5 – a concept I will explain in the next section. To conclude this section, the process of situating my research involves a dual search for my influence on the informants, as well as a consideration of my own discursive understanding; they are both important focuses in the analysis. This means that throughout this project, I will continuously be situating my reading of the contextualised empirical material. I will be both interpreter and interpreted. The results of this process will first be available when the process has been 'frozen' in the text of a dissertation. I will leave this discussion here, and turn to the theoretical foundation and analytical tools I use, starting with how technology research provides important perspectives in my work, before I proceed with a discussion of how I treat gender as an analytical concept.

Technology research The most important analytical perspective in this project is the focus on gender, but studying the use of technology also requires some notion of technology, not only as metal and wires, but also of its own social meaning. Questions concerning how technology and society can best be understood together have been important within both mainstream and feminist technology research at least the last three decades. The feminist research on technology6 has been developed theoretically along with mainstream technology research and it is possible to understand this development as three subsequent waves implying different attitudes towards technology.7 My own project can be placed within the third wave of technology research, starting in the early 90s, with a social constructivist notion of technology in which technology and society is seen as constructing each other in a mutual process. The focus of interest has turned away from the earlier interest in innovation and development of technology towards a new interest in dispersion and use, with an increasing attention to how users of technological artefacts apply them with a great portion of freedom. There are several theoretical perspectives developed under the umbrella of social constructivism8 with a wish to avoid the technological

The multi-dimensional stories of the gendered users of ICT

determinist trap on one side, and the social determinist on the other, and at the same time to grasp the complex relations between different social and technical factors which affect and shape each other. The perspective which has had most influence on my own thinking about the relationship between technology and society is Silverstone, Hirsch and Morley's theory of domestication.9 The concept of domestication has traditionally been applied to the integration of animals into households, a process during which both the animals and the household changed character and adapted to each other. The concept of domestication has been used by Silverstone et al. to refer to a similar process regarding technology; a process in which technological artefacts are introduced to a household's 'moral economy'. They see this process as a continuous development, including several steps or phases,10 which address how people acquire, adopt, get used to, learn to use and how they present their relations to technological artefacts.11 The notion of an existing 'moral economy' within households, reminds us that a household (or an individual) is not an empty space to be filled by the presence of technology; it already has certain structures, meanings, and routines, and it is within this framework technological artefacts are being domesticated. I am not using the theory of domestication as a methodological foundation, but I think that this theory has an important function in a study that seeks to reach an understanding of how people create their relations to computers. Therefore I use the theory of domestication as a fundament within the theoretical frames of the project; as a constant reminder of how the relationship between people and computers should be seen as a continuous process of mutually interaction and shaping. When I am referring to the 'shaping' of technology, I am not aiming at a physical shaping, but to the way the meaning of both gender and technology are affected in a people-technology relationship. A technological artefact should not be treated only as a material expression of a discourse. It is also in itself an active articulating participant in a discourse – surrounded by meaning.12 One of the informants illustrates technology's ability to express meaning when she says 'The computer milieu is masculine in the sense that it is a room with computers, printers and wires everywhere.' In my project, I am focusing on computing, which includes a conglomerate of physical equipment – hardware as well as software. It includes the notion of knowledge, as well as the 'perceived meaning of computing', which points to general understandings and stereotypi-

167

168

Researching ICTs in Context

cal ideas of what computing is.13 Gender has throughout the history of computing been one important factor within these understandings. Even though gender is a relational phenomenon, it has often been treated as synonymous with 'women', not only in technology research, but also more general within feminist research (and elsewhere). It is primarily in the third wave of technology research that both men and women's relations to the computer have been studied.14 Both men and women are included in my study too, and I agree with the viewpoint that we have to understand how both men and women create their relations to computers if we want to understand how gender makes a difference in relation to computers. Gender as a relational phenomenon is based on a definition of the borders between the genders, which makes it necessary to have knowledge about both sides (Blom 1994: 50). But what exactly is gender? And how do we 'find' it within the empirical material?15

Gender theory inspired by Joan Scott In accordance with the social constructivist perspective that I have on technology, I also employ a constructivist notion of gender, based on the historian Joan Scott's post-structuralist theory.16 Scott points to the importance of seeing gender as a social construction, and her often quoted definition states that «gender is a constitutive element of social relationships based on perceived differences between the sexes.' (Scott 1988: 42). It does not have to include 'fixed and natural physical differences' between women and men, but is 'the knowledge that establishes meanings for bodily differences' (Scott 1988: 2).17 I agree with Scott that gender should be seen as a historical and social construction, which explains and gives norms, rules, and directions to men and women living in society. Furthermore, it is important to distinguish 'gender' from the real instances of men and women. In this project I use 'gender' when I talk about the discourses, the structure of thoughts that provides the guidelines to the 'real instances'. However, it is the living men and women that can teach us more about 'what it means to be a man or a woman'; how men and women experience their situation in a society where gender have acquired meaning also in seemingly non-gendered situations and spheres.18 I think that Toril Moi's reading of Simone de Beauvoir illustrates my understanding of gender very nicely when she writes: 'For Beauvoir, a

The multi-dimensional stories of the gendered users of ICT

woman defines herself through the way she lives her embodied situation in the world, or in other words, through the way in which she makes something of what the world makes of her.' (Moi 1999: 72).19 This quote illustrates how I work with gender, as a discursive category constructed around perceived differences between men and women, as something we meet in the world, something that is already there. It is through the ways we deal with it that 'living in the world as men and women' acquires its meaning. In this way, my initial question: 'in which ways does gender affect men and women's relations to computing?' can actually be spelled out as two questions. The first is 'How is gender perceived in relation to computing?' That is, in which ways do men and women think that gender makes a difference in computing? The second is an analytical question concerning the effects of gender: 'How do men and women deal with the perceived gender?' How are they combining the meaning of gender in relation to computing with other discursive positions they possess? The first question presupposes that gender already exists and is perceivable to individuals. The second question presupposes that individuals have the power to deal with the perceived gender in different ways – that gender is not a fixed structure, but supports different solutions of how to be a man or a woman. My analysis must therefore search for how individuals both perceive and deal with gender. Even though I find Joan Scott's writings inspiring for my own work, she does not offer many operational concepts or strategies for a practical analysis. In my search for an analytical method, I found the political philosophers Ernesto Laclau and Chantal Mouffe's theory of discourse interesting.20 Scott, Laclau, and Mouffe have a common point of reference in the tradition following Jacques Derrida and Michel Foucault, and they share the basic theoretical framework of post-structuralism, with an emphasize on social structures, institutions, categories etc. as socially constructed, as I have illustrated through Scott's notion of gender. Furthermore, they share the rejection of the existence of an original 'essence', or 'truth' behind history as we can observe it, and the study of history is proposed as a study of movements of meaning in a historical context; movements that are consequences of other movements in an endless line of signification (cf. Scott 1988: 4, Laclau & Mouffe 1985, 105f, 111-112). And it is within this framework Laclau and Mouffe offer some operational concepts or strategies for a practical analysis in their theory of discourse, which I will discuss below.

169

170

Researching ICTs in Context

Discourse theory as an analytical tool Laclau and Mouffe's theory of discourse21 is only one among a broad variety of discourse analytical propositions. Discourse analysis has its roots within the field of linguistics, but is being used today in various fields, in different ways, and not limited to either text or language, which is the case of Laclau and Mouffe's theory. They suggest that we see politics, or the social construction of meaning, as a struggle over the power to formulate the hegemonic (dominating) discourses. They emphasise the importance of seeing everything through the logic of discourse, as our only possible way to make sense of the world is through our discursive understanding, a point I fully agree with. Within different discourse analytical methods, there are, however, different things which are being recognised as articulations and thereby possible to analyse with a discourse theoretical approach. The traditional, though increasingly abandoned, view of textuality as concerning written text, does not include the materiality of technology in its analytical focus (cf. Derrida 1967/1997. See also Vehviläinen 1997: 9, Hansen 2000: 140). The materiality of technology refers to Bruno Latour's sociology of technology, where he has emphasized the importance of including technology as 'the missing masses' in social sciences, due to the material technology's influence on people's actions (Latour 1992, Cf. Lie 1998: 25). I share the belief that both physical objects as well as individuals' actions are expressions of meaning, and the study of discursively produced meanings should include all forms of production of meaning. In the next section I discuss how meaning is produced within a discourse and present the concept of 'subject position', which is the most important analytical concept I employ from discourse theory.

The logic of discourse The key concept within discourse theory is discourse, which here is defined as a dominating or hegemonic meaning within one area. I use the concept of discourse to talk about limited areas, like the discourse about computing, the discourse about gender equality etc.22 A discourse is perceived as a fixed objectivity through 'homogenisation of an interior' and 'exclusion of an exterior'. The homogenisation of the interior means that all the meaningful components which together create a discourse have to let

The multi-dimensional stories of the gendered users of ICT

go of parts of themselves in order to become coherent with the other meaningful components. Through the exclusion, a discourse produces a surplus of meaning, and this surplus is confined to the field of discursivity, a concept with which Laclau and Mouffe want to remind us that everything is discursively constructed, and included in the logic of discourse.23 Together, these two forces in the logic of a discourse create an impression of a coherent internal meaning, and an unquestionable discourse. The impression of objectivity also covers the dynamic and temporary character of a discourse (cf. Laclau & Mouffe 1985, chapter 3), a dynamics partly due to the absence of a fixed centre in the continually moving discourse, which implies that meaning is always under negotiation and construction ( Cf. Torfing 1999: 39ff). Articulations, the expression of meaning, will always involve something new, and is therefore potentially producing new meaning (cf. Laclau & Mouffe 1985: 105).24 To draw the exact limits of a discourse is impossible, and the way Laclau and Mouffe solve this problem is to refer to 'the radical otherness' – that which is definitely not a part of a discourse, or that which has nothing in common with the interior of the discourse (cf. Laclau 1995).25 There are some common misreadings26 of post-structuralist theories including discourse theory,27 where one of the basic arguments about the social construction (the claim that social institutions, structure, categories etc. should be seen as socially constructed and that there is no necessary links between an essence, origin, or a starting point and the outcome) has often been misread as claiming that reality does not exist (or that material reality does not exist). The post-structuralist argument is not claiming that reality is either false or non-existent. Rather, it says something about how reality has come to look like it does. Furthermore, emphasis on the contingent and dynamic character of a discourse has been misread as a statement that 'anything is possible'. The rejection of essence means that the movements of meaning are dynamic and can potentially move in any direction. This does not mean that the movement of meaning is entirely free: it is always connected to and directed by a context in which certain possibilities are revealed while others remain hidden. At the same time, every new movement of meaning becomes a part of the context itself, which means that the articulating individuals are created by, as well as creators of, the social reality they live in.

171

172

Researching ICTs in Context

Subject position – a discursive point of identification A subject position is a discursive point of identification within a discourse. It is through identification with available subject positions that we construct our identity. A subject position defines the possible frames of action of the individual in a specific context. We all have several subject positions from different discourses, and they are of varying importance in different situations. Subject positions are more or less easy to acquire or to avoid for different people, as there are some we cannot avoid (like being a mother, sister etc.), some we cannot reach, and a whole spectre in between. It is possible to negotiate a subject position, but still the way we activate a subject position has to be recognisable within the existing frames in order to be accepted. Gender is an important category in my project, and gender can be activated as the most important subject position in itself, but it can also act as an important marker of seemingly non-gendered subject positions. Recall Simone de Beauvoir's emphasis on the interaction between individuals and the society: 'a woman is what she makes of what the world makes of her';28 the discursively defined subject positions are what 'she' is offered from the world – it is the starting point she is given by the discourse, and it is her perception and treatment of this starting point that makes her a woman. The subject position is perhaps the most important concept in my analysis, as it deals in particular with alternative discursively defined points of identities which individuals can relate to. It provides a concept with which to analyse how discourses have importance to individuals as well as how individuals have the possibilities of negotiating with discourses, a point I will illustrate below. To recapture the logic of a discourse, is to acknowledge that it operates through the exclusion of all 'other possibilities', while the included meaningful parts are homogenised in order to create an internal consistency, as well as to make any internal inconsistency become invisible. A socially constructed meaning can thereby be presented as 'natural' and 'given'. To use the theory of discourse as a method means exploring how the meaningful elements of a discourse are tied together in specific ways in order to produce a specific meaning. It is the search for the possible, as well as the impossible, frames surrounding social actions in a given context.29 One of the strengths of discourse theory is its focus

The multi-dimensional stories of the gendered users of ICT

on the interplay between people and discourses, in my case this is the interplay between the students and discourses about computing.

Looking for gender in individuals' relations to computers I will present two of the informants in my study, a man and a woman, in order to illustrate how discourse theory works as an analytical tool in my search for the meaning of gender and of computing. In this presentation I will concentrate on how a discourse produces subject positions. The intention of the discussion is purely to illustrate how I work with gender, and not to display what gender means, as other representatives of the informants could – and would have – displayed other 'solutions'. We will start by looking at the practical side of their relation to computers, that is, practical tasks and activities at the computer. The two informants both started with playing computer games, they advanced to schoolwork and more 'serious' writing, to surfing the Internet and to using e-mail. They both have technical experience with the physical computer, the man by building his own computer, the woman by opening up and fixing hers, and they both take care of the software side and manage the operating system. They have been engaged in rather similar activities since their first meeting with computers. The most striking difference is that the man was introduced to computers as a child, while the woman was introduced to computers as an adult. Except for the fact that the first person is a man, and the second is a woman, what role does gender play in their relations to the computer? Are any of these tasks perceived as more or less masculine or feminine? Earlier research has pointed to a number of gendered differences related to computers, such as computer games being more popular among boys, girls' delight in communicating through e-mail or searching for useful information on the net, and so on (see e.g. Håpnes & Rasmussen 1997). These are differences recognised by earlier research, and some of these gender differences also existed in the selfpresentations of the men and women in my preliminary study at Department of Humanistic Informatics in 1997, as a tendency among women to express their computer knowledge in terms of insecurity, and a tendency among the men to express theirs in terms of self-confidence.

173

174

Researching ICTs in Context

A similar mismatch is found in several other research projects, both in Norway (KUF 2001: 2, Frølich et al. 2002: 9) and in US (Freshman Norms 2002). However, this indicates that men's and women's practices in relation to computers might have changed more than conceptions about what gender means in relation to computers. These conceptions are often repeated as given, as 'truths' about how gender makes a difference in computing, both by researchers (see for instance Nissen 1996), planners of gender equality strategies (cf. KUF 2001), and the male and female computer users themselves (cf. the presented mismatch). It seems to be time to explore further some of these 'truths' about how men and women relate to computers.

Illustrating the method We shall first be looking through the eyes of the male student. In an email he tells me that: When I went to junior secondary school and perhaps in secondary school, I could see a difference between boys and girls in relation to computing. The boys had a much bigger interest. But now, we can see that girls have got a new interest for computing. … The reason could be that everyone is expected to know a little bit about computing in today's society. […] … as a boy, I have been engaged in computing ever since I went to primary school.

And in another e-mail: In my circle, girls have always been a little bit 'interested' in computing. Perhaps not as much as us boys.

In this articulation the male student combines 'boy' and 'interest in computing', while the girls had some kind of interest – in quotation marks – as if it wasn't a real interest. He creates a discourse that establishes the connection between boys and computing, with girls partly left on the side. 'Girls' are in a process of moving from a position on the outside to a position on the inside in the discourse of computing, but they do not seem to be entirely integrated in the discourse yet, as the difference is still there. The girls seem to have a different objective from the boys in their 'new interest for computing'; while the boys have a genuine interest for the computer itself, girls have an interest in participating in a social setting where ICT has become increasingly important. Despite this movement, the most obviously avail-

The multi-dimensional stories of the gendered users of ICT

able subject position in this discourse (about 'interest in computing') continues to be associated with the masculine gender, and gender is still a personal factor that tells us whether a person has an immediate access to the subject position of 'computer interested' or not. And it is 'as a boy' the student himself acquires a subject position in relation to computing.30 Based on his own gender, this male student has got access to a masculine subject position. But what about the female student? She writes: Because I don't have a brother, my sister and I had to fill that 'gap' by learning practical tasks that traditionally often are performed by men.

And she goes on to talk about how this gives her an advantage in relation to computing. Otherwise I am a fearless type of person … I think that all of this is an advantage for working with computing-

She is accepting a discourse about the connection between masculinity and computing, and in a discourse that establishes the masculine gender as a vital part, the feminine gender will be excluded as a 'radical otherness'.31 In order to acquire a subject position within that discourse, girls and women will have to find other solutions, as the female student illustrates. She inscribes herself in the masculine subject position by arguing that she is used to doing a man's work. She is used to taking on masculine subject positions. If we go back to the two questions I asked earlier, 'how is gender perceived in relation to computing', and 'how do men and women deal with the perceived gender', we can see how these two students have both perceived gender as something that 'makes a difference within computing'. They both deal with the perceived gender. In different ways though, as the male student can inscribe himself in the masculine discourse about computing without any other explanations than the fact that he is a boy, and that boys in his generation were more interested in computing than girls were. The female student, on the other hand, has to add one important feature to qualify, which is a general masculine practical competence – before she inscribes herself in the available masculine subject position. However, by doing that, she is also doing something with the subject position within computing as she defines this position as based on the tradition of upbringing which trains children in different tasks depending on the child's gender, and she denies any connections between computing and gender as essential or necessary differences between people. Gender is simply about social training.

175

176

Researching ICTs in Context

Both of them also use their understanding of gender in relation to computing when they consider their fellow students; the man directs the attention to his own observation of the women, and he claims that they have computer competence (the female students 'have a lot of knowledge about what they are doing'). The woman, on the other side, directs the attention to the observation that the men did not have any more knowledge then she had ('..the men who were present did not know more than I did …'). Both seem to be starting from within a discourse of the masculinity of computing, and it is the failure of this discourse they comment on; gender does not seem to make the difference they expected. This also illustrates how they both seem to enter the world of computing from different angles – the man might have thought he would be in the top league as a man among women, while she seems to have expected to be in an inferior position as a woman among men. The observations are formulated with reference to their idea of gender and computing, and it illustrates how our perception of gender tells us what to expect from others. This analysis illustrates how gender already exists as ideas – not meaning men and women, but rather as potentially masculine and potentially feminine associations and subject positions.32 In this case, we see gender as cultural assumptions of what kinds of relations to the computer it is likely we think men and women have. But, we also know that this does not mean that men and women will necessarily match with the subject positions. The analysis illustrates how individuals need strategies and arguments either to make themselves fit into the ideas or categories they perceive, or to change the expectations. If we want to disclose some of the stereotypical ideas about gender and computing, I think it is important to look for the different, and perhaps the oppositional voices, those who break the stereotypical ideas, like that of one female informant who is 'in love with her machine', and the male informant who hates games. And we need to be aware of the multidimensional stories within individuals who combine their own 'moral economy'33 with the discourses they are continually confronted with in the world.34 This is where the strengths of Joan Scott's insistence on contextualising fit together with the theory of discourse's ability to point to how some possibilities become visible while others remain hidden, or are perceived as impossibilities; it points to how some meanings have been suppressed in order to give the chaotic reality a coherent meaning. This could be thought of as an exploration of the

The multi-dimensional stories of the gendered users of ICT

'failures' in history as much as the 'successes' – failures to impose, to make a difference, to leave a trace, or a hegemonic discourse. The failure to make a difference seems to be one of women's problems in history, as for instance in the case of the computer. The fact that women have been a rather large group of computer users from the personal computer's early history until today, and the fact that one of the earliest tasks of the computer was to relieve the female secretary's typewriter, has not really made any difference in the perception of the computer as a masculine artefact. Another important feature of the theory of discourse is its ability to reveal how individuals' actions and statements are produced within the frame of possibilities – which frames that are perceived by whom in which contexts. This method provides tools for exploring what individuals do with the reality they live in. It has the ability to explain similarities and differences among individuals. Even more important, it has the ability to explain inconsistencies within the individual that at first sight might look like paradoxes or contradictions, as well as to explain how one person can 'live out' these paradoxes without experiencing herself as incoherent. In other words, discourse theory can help to explore the multitude of discourses each individual confronts and the multidimensional stories these individuals tell. It is these multidimensional stories I am studying in my Ph.D. project. I think that the mismatch between people's practical abilities and self-presentation that I mentioned in the beginning is connected to the way the hegemonic discourses about computing still seems to favour men. I also think that some of the stereotypical ideas about gender and computing can be replaced by more detailed descriptions of individuals' relations to computers. Perhaps, too, some of these descriptions can serve as the basis of new subject positions within computing, to men and women who cannot find any suitable positions among the stereotypical ones we already know.

References Asdal, K., Brenna, B. & Moser, I. 2001. 'Introduksjon: teknovitenskapelige kulturer'. In Asdal, K., Brenna, B. & Moser, I. (eds) Teknovitenskapelige kulturer. Oslo: Spartacus Forlag AS. 9-90. Aune, M. 1996. 'The computer in everyday life: patterns of domestication of a new technology', in Lie, M. & Sørensen, K.H. (eds) 1996. Making Technology Our Own? Oslo: Scandinavian University Press. 91-120.

177

178

Researching ICTs in Context

Barker, J. & Downing, H. 1980. 'Word processing and the transformation of patriarchal relations of control in the office'. In MacKenzie, D. & Wajcman, J. (eds) The Social Shaping of Technology. Buckingham: Open University Press. 147-164. Berg, A.-J. 1996. Digital Feminism. STS-rapport nr.28, dr.polit.avhandling. Institutt for sosiologi og statsvitenskap, NTNU: Trondheim. Berg, A.-J. 1998. 'Det beskjedne vitne'. In Asdal, K., Berg, A.-J., Brenna, B., Moser, I. & Rustad, L. M. (eds) Betatt av viten – bruksanvisninger til Donna Haraway. Oslo: Spartacus Forlag A/S. 76-79. Blom, I. 1994. Det er forskjell på folk – nå som før. Om kjønn og andre kriterier for sosial differensiering. Oslo: Universitetsforlaget. Braidotti, R. 1994. Nomadic Subjects; Embodiment and Sexual Difference in Contemporary Feminist Theory. New York: Columbia University Press. Cockburn, C. 1985. Machinery of Dominance: Women, Men, and Technical Knowhow. London: Pluto Press. Corneliussen, H. 1997. 'Myten om kvinner og data: en teoretisk tilnærming til studiet av kjønn og IT'. Semesteroppgave. Universitetet I Bergen: Humanistisk informatikk. Derrida, J. 1967/1997. Of Grammatology. Baltimore and London: The Johns Hopkins University Press. Dyrberg, T.B., Dreyer Hansen, A. & Torfing, J. (eds) 2000. Diskursteorien på arbejde. Frederiksberg: Roskilde Universitets Forlag. Fairclough, N. & Wodak, R. 1997. 'Critical Discourse Analysis'. In van Dijk, T.A. (ed.) Discourse as Social Interaction. London: Sage. 258-284. Faulkner, W. 2000. 'The power and pleasure? A research agenda for "making gender stick" to engineers'. Science, Technology, & Human Values, Vol. 25, No. 1. 87-119. Foucault, M. 1971/1999. Diskursens orden. Oslo: Spartacus Forlag A/S. Freshman Norms Report. 2002. [cited 2 February 2002]. At: http://www.gseis.ucla.edu/heri/00_exec_summary.htm Frølich, T. & Erstad, O. 2002. Omstillingens utgangspunkt: En kartleggingsstudie av PILOTer våren 2001. PILOT: Prosjekt Innovasjon i Læring, Organisasjon og Teknologi. University of Oslo: ITU. Grint, K. & Gill, R. 1995. 'The gender-technology relation: contemporary theory and research'. In Grint, K. & Gill, R. (eds) The Gender-Technology Relation. Taylor & Francis: London, 1-28.

The multi-dimensional stories of the gendered users of ICT

Hansen, A.D. 2000. 'Lokaludvalg som konstruktion af lokale politiske fællesskaber'. In Dyrberg, T.B., Dreyer Hansen, A. & Torfing, J. (eds) Diskursteorien på arbejde. Frederiksberg: Roskilde Universitetsforlag. 131-160. Haraway, D. 1997. Modest_Witness@Second_Millennium.Female– Man_Meets_OncoMouse. Feminism and Technoscience. New York: Routledge. Haraway, D. 1991. 'Situated knowledges: the science question in feminism and the privilege of partial perspective', in Haraway, D. Simians, Cyborgs, and Women. The Reinvention of Nature. London: Free Association Books. 183201. Harding, S. 1991. Whose Science? Whose Knowledge? Thinking from Women's Lives. Ithaca NY: Cornell University Press. Håpnes, T. 1992. 'Hvordan forstå mannsdominansen i datafaget? En dekonstruksjon av fag- og kjønnskultur', in Annfelt, T. & Imsen, G. (eds) Utdanningskultur og kjønn. NTNU: Skriftserie ved Senter for kvinneforskning. 3/92. 155-183. Håpnes, T. & Rasmussen, R. 1997. Internett – jentenett? Ungdomsskolejenters databruk og datainteresse. NTNU: Skriftserie ved Senter for kvinneforskning. 7/97. Jørgensen, M.W. & Phillips, L. 1999. Diskursanalyse som teori og metode. Frederiksberg: Roskilde Universitetsforlag. Kjørup, S. 1996/1999. Menneskevidenskaberne: Problemer og traditioner i humanioras videnskabsteori. Frederiksberg: Roskilde Universitetsforlag. KUF. 2001. 'Dilla på data', Likestilt no. 2. Kirke- utdannings- og forskningsdepartementet/Ministry of Church Affairs, Education and Research. [cited 15 February 2002]. At: http://odin.dep.no/archive/kufbilder/01/05/likes011.pdf Laclau, E. & Mouffe, C. 1985/1994. Hegemony and Socialist Strategy. Towards a Radical Democratic Politics. London: Verso. Laclau, E. 1995. 'Subject of politics, politics of the subject'. Differences: A Journal of Feminist Cultural Studies, Vol. 7. No. 1. 146-164. Latour, B. 1992. 'Where are the missing masses? The sociology of a few mundane artifacts'. In Bijker, W. & Law, J. (eds) Shaping Technology/Building Society. Studies in sociotechnical change. Cambridge MA: The MIT Press. 225-258. Latour, B. 1991. 'Technology is society made durable', in Law, J. (ed.) A Sociology of Monsters. London: Routledge. 103-131.

179

180

Researching ICTs in Context

Law, J. 1993. 'Technology and the heterogeneous engineering: the case of Portuguese expansion', in Bijker, W., Hughes, T.P. & Pinch, T. (eds) The Social Construction of Technological Systems. Cambridge MA: The MIT Press. 111-134. Lie, M. 1998. Computer Dialogs: Technology, Gender and Change. NTNU: Skriftserie ved Senter for kvinneforskning. 2/98. Lie, M. & Sørensen, K.H. 1996. 'Making technology our own? domesticating technology into everyday life'. In Lie, M. & Sørensen, K.H. (eds) Making Technology Our Own? Oslo: Scandinavian University Press. 1-30. Mellström, U. 1996. 'Teknologi och maskulinitet: Män och dera maskiner'. In Sundin, E. and & Berner, B. (eds) Från symaskin till cyborg: Genus, teknik och social förändring. Stockholm: Nerenius & Santerus Förlag. 113-139. Moi, T. 1999. What is a Woman? And Other Essays. Oxford: Oxford University Press. Nissen, J. 1996. 'Det är klart att det är grabbar som håller på med datorer! Men varför är det så?' In Sundin, E. & Berner, B. (eds) Från symaskin till cyborg: Genus, teknik och social förändring. Stockholm: Nerenius & Santerus Förlag. 141-161. Noble, D. 1984. Forces of Production. New York: Alfred Knopf. Pinch, T. J. & Bijker, W.E. 1993. 'The social construction of facts and artifacts: or how the sociology of science and the sociology of technology might benefit each other'. In Bijker, W.E., Hughes, T.P. & Pinch, T. (eds) The Social Construction of Technological Systems. Cambridge MA: The MIT Press. 17-50. Rosenbeck, B. 1992/1996. Kroppens politik: Om køn, kultur og videnskab. København: Museum Tusculanums Forlag. Rustad, L.M. 1998. 'Kunnskap som delvis forbindelse'. In Asdal, K., Berg, A.-J., Brenna, B., Moser, I. & Rustad, L. M. (eds) Betatt av viten – bruksanvisninger til Donna Haraway. Oslo: Spartacus Forlag A/S. 120-144. Scott, J.W. 1988. Gender and the Politics of History. New York: Columbia University Press. Silverstone, R., Hirsch, E. & Morley, D. 1992. 'Information and communication technologies and the moral economy of the household'. In Silverstone, R. & Hirsch, E. (eds) Consuming Technologies: Media and information in domestic spaces. London & New York: Routledge. 15-31. Søndergaard, D.M. 2000. 'Destabiliserende diskursanalyse: veje ind i poststrukturalistisk inspireret empirisk forskning'. In Haavind, H. (ed.) Kjønn og fortolkende metode. Metodiske muligheter i kvalitativ forskning. Oslo: Gyldendal Norsk Forlag. 60-104.

The multi-dimensional stories of the gendered users of ICT

Sørensen, K. H. 1997. 'Learning technology, constructing culture. Sociotechnical change as social learning'. Unpublished SLIM-paper. Trondheim: STS, NTNU. Torfing, J. 1999. New Theories of Discourse - Laclau, Mouffe and Zizek. Oxford: Blackwell Publishers. Vehviläinen, M. 1997. Gender, Expertise and Information Technology. Academic Dissertation. Tampere: Department of Computer Science, University of Tampere. Webster, J. 1995. 'What do we know about gender and information technology at work? A discussion of selected feminist research'. The European Journal of Women's Studies, Vol. 2, No. 3. 315-334. Winner, L. 1980. 'Do artefacts have politics?' In MacKenzie, D. & Wajcman, J. (eds) 1985. The Social Shaping of Technology. Milton Keynes: Open University Press. 26-38.

Notes 1

Even though it would have been preferable to have an equal number of men and women in my study (cf. p. 168), which was impossible to attain within one class due to 65% women among our students, this does not cause any problems to my project, as it is not aimed at comparing men with women, but rather to study individuals' gender strategies in the (gendered) field of computing. 2 Sandra Harding discusses this in particular in Whose science? Whose knowledge? (Harding 1991). 3 Cf. the section of this chapter entitled 'Gender theory inspired by Joan Scott'. 4 One of the most important messages is that of programming as a skill that can be acquired through practice (like learning to ski – theory is important, but useless without practical training). 5 See paragraph containing note 4. 6 For a more detailed overview of feminist technology research, see e.g. Grint & Gill (1995), Webster (1995), Berg (1996), Lie (1998), Faulkner (2000). 7 See for instance Webster (1995), Grint & Gill (1995). These three waves represent one of the common ways of classifying this development, starting with the 'impact' perspective of the 70s, where technology primarily was perceived as determining changes in society and not the other way around. See e.g. Barker & Downing (1980); cf. Lie (1998: 19-20.) The most important claim made within the second wave, the 'social shaping' tradition of the 80s, is that technology alone does not change society based on any technological necessities. Instead, the determining factors are to be found in social, political, or eco-

181

182

Researching ICTs in Context

nomical assessments in the development of technology. See for instance Winner (1980), Noble (1984), Cockburn (1985.) 8 For instance Bruno Latour (1991) and Law (1993) on actor-network theories, Pinch & Bijker (1993) (SCOT: Social Construction of Technology). 9 Silverstone, Hirsh & Morley (1992). For a discussion of this theory, see also Lie & Sørensen (1996), Berg (1996). 10 The four phases are appropriation, objectification, incorporation and conversion. 11 Knut H. Sørensen has abandoned the four phases and suggests the process to be seen in three levels – a practical level considering what people do, a symbolic level focusing on the production of meaning, and a cognitive level that focus on the process of learning to use technology (Sørensen 1997). 12 Informant F1. 13 I use the words computer and computing alternately, because the Norwegian word 'data' can refer to both. 14 See for instance Håpnes (1992), Aune (1996), Mellström (1996), Nissen (1996), Lie (1998), Faulkner (2000). 15 How to carry out an empirical investigation about gender is itself a difficult question. How can we for instance stimulate people to give other answers to questions about gender than the 'politically correct' idea that gender should not mean a difference in the Scandinavian context of equality? These questions will be treated elsewhere, while this presentation will focus on how to analyse gender in an already existing empirical material. 16 Gender is a concept first used by Gayle Rubin in her essay 'The Traffic in Women' from 1975, in order to reject biological determinism. Rubin used 'sex' as a reference to the body, while gender referred to social norms that were responsible for explaining bodily differences. (Moi 1999: 24). 17 Although Scott includes the bodily experience in her gender concept, «nothing about the body, including women’s reproductive organs, determines univocally how social divisions will be shaped.» (Scott 1988: 2) Scott is thereby not among the poststructuralist inspired feminists which reject the physical body’s existence or the physical experience, as Toril Moi has criticised poststructuralists for (see Moi 1999, chapter 1), but she claims that gender is solely a social construction. 18 See for instance Scott's discussion of gender in relation to the working-class history in Scott (1988: chapter 3). 19 Moi's reference is to de Beauvoir 1948. The Second Sex. 20 Laclau and Mouffe's discourse theory (See for instance Laclau & Mouffe 1985) should not be confused with other types of discourse analysis. (See e.g. Jørgensen and Phillips 1999.) It is a theory originally evolved from criticism of other theories and through political debates more than from empirical work. However, more empirically based research projects inspired by Laclau and

The multi-dimensional stories of the gendered users of ICT

Mouffe's discourse theory are reported in Dyrberg et al. (2000), which gives an account of a varied and productive exploration of the theory of discourse among Danish researchers during the last decade. For a historical and theoretical review of discourse theory, see Torfing (1999). Torfing is one of the Danish researchers who has greatly influenced my own exploration and adoption of the discourse theory. 21 My use of discourse theory as an analytical method can seem to be confusing due to the word 'theory' involved in its name. Using discourse theory as a method is also a question of different opinions (cf. Torfing 1999: 12), and my use of the theory of discourse is not a copy of Laclau and Mouffe's use, as they are dealing with a very different kind of material from that I am working with. This discussion does not consider every aspect of either their theory or my use of it as a method, but is limited to the most important aspects and concepts in my work. 22 The concept of discourse is used in quite different ways, like one of Michel Foucault's (1971) ways of defining discourse as meaning the whole social production of meaning in any time in history or Norman Fairclough's limited concept, which only includes some social spheres, while other spheres (i.e. economy) have to be considered through other concepts (cf. Fairclough 1997.) 23 Meaning which in one discourse is confined to the field of discursivity is a part of another discourse (or several discourses). The field of discursivity is both the presupposition for a discourse to reach a coherent interior: the possibility to homogenise the included parts, as well as a threat, as the possibility of including meaning from the field of discursivity will always threaten to overthrow the discourse. (Laclau & Mouffe 1985: 111). 24 Laclau and Mouffe (1985: 105) call '…articulation any practice establishing a relation among elements such that their identity is modified as a result of the articulatory practice.' (cf. Torfing 1999: 101). 25 Jacob Torfing writes that a 'discourse, or a discursive formation, establishes its limits by means of excluding a radical otherness that has no common measure with the differential system from which it is excluded …'. (Torfing 1999: 124). 26 Rejection of these kinds of misreadings can be found in most books dealing with post-structuralist theories, see for instance Torfing (1999), Dyrberg et al. (2000), Phillips and Jørgensen (1999), Søndergaard (2000: 68ff), all of which have been influential in my own work. 27 Laclau and Mouffe's argument against the misreading that their theory includes a denial of the existence of reality, is that '[w]hat is denied is not that such objects [of the physical world] exist externally to thought, but the rather different assertion that they could constitute themselves as objects outside any discursive condition of emergence.' (Laclau & Mouffe 1985: 108.) A phenomenon exists without a discourse, but it cannot be understood without a context, which always will bring in a discursive basis for interpretation. 28 Cf. my earlier arguments earlier in the chapter on feminism and technology.

183

184

Researching ICTs in Context 29

I am of course also a part of the same society that the informants are, and my ability to see the ‘impossible’ (or the excluded meaning) is to a certain degree limited. 30 The articulations in these examples are taken from answers the informants gave on questions concerning gender, which can make it understandable that they answer in terms of gender, but this is not necessarily the case, as other informants' answers can demonstrate; the interesting question is not whether the answers bring in gender, but how they bring in gender (or not), and articulations to the questions about gender still reveal how they deal with the meaning they perceive in relation to the questions. This discussion will be completed in my forthcoming dissertation. 31 Cf. the sentence linked with note 25. 32 The division between on one hand ideas, categories and mental structures (femininity/masculinity) and lived lives (women/men) on the other, represents a fundamental perspective within research which want to avoid essensialising gender. (See for instance Rosenbeck 1992.) This division makes it possible to see gender as a social construction. It also makes it possible to see masculinity and femininity as something found simultaneously in one individual. 33 See the point linked with note 4. 34 Cf. Braidotti's 'nomadic epistemology' in which she makes a distinction between 'differences between men and women,', 'differences among women,' and 'differences within each woman'. (Braidotti 1994: 158).

9 Seduced by numbers? Helen Jøsok Gansmo

How and why is research on girls and computing applied in decision making? Girls' apparent lack of interest in computers has been regarded as a public problem in Norway for about 20 years and thus been subjected to thorough research. Such research is often funded — as it the communication of its results to politicians and policymakers — on the notion that that research results and knowledge are the basis for the best possible decisions through which to solve this problem. Social science makes use of both qualitative and quantitative methods to investigate social problems. Quantitative research methods can for instance help illuminate a problem by detecting serious and important biases. Such figures are informative, very concrete and easy to distribute to politicians, media and the public at large. The figures are often perceived as solid facts that show that boys are far more interested in computing than are girls. Qualitative research on the other hand does not focus on quantities, but tries to illuminate unique features of the matter in question. Rich material on for instance why and how different the teens actually make use of the computer is gathered through long interviews with boys and girls regarding their own (lacking) experience with and their ideas of computers and computer users. This chapter will focus on how research results regarding girls and computing is interpreted and applied in policy making. I first describe some of the available research findings regarding girls and computing, before I make a theoretical approach to how and why 185

186

Researching ICTs in Context

(not) research results are applied in policy making and what kind of research is perceived to be reliable and trustworthy. Finally I discuss on the interpretations of the research results on the part of the decision makers and the choices these seem to lead to. I will show what policymakers and politicians regard as their knowledge based on girls and computing, and how they more or less systematically diminish qualitative research results or narratives as not useful or scientifically reliable.

Available research results on girls and com computing The numbers point in the same direction

In 1990 in Norway, 44% of the students in secondary schools had access to computers at home, 56% of boys and 32% of girls (IBM in St. meld. 24). Results from a quantitative survey conducted by SSB (Central Bureau of Statistics) in 1995 showed that in secondary schools 62% of boys and 52% of girls had access to a computer at home. Over 50% of these boys claimed that they are the most frequent users, while only 20% of the girls claimed they use the home computer the most (IT i norsk utdanning. Plan for 1996-99). This is interpreted as a considerable difference between boys and girls regarding interest for and use of IT. In 1999, the number of students per available computer in secondary schools in Norway is 8,9. This is relatively higher than in other countries, but equipment is not up-to-date (SITES 1999 in IKT i norsk utdanning. Plan for 2000-03). Half of teachers have participated in computing courses in the use of spread sheets, the Internet and word processing. Under 20% of teachers have participated in courses on how to implement computing as a teaching aid (Fafo 1999 in IKT i norsk utdanning. Plan for 2000-03). The numbers showing teachers' lack of computing knowledge and girls' lack of interest and access to computers have been relatively stable over the 20 year period in which girls and computing has been regarded as a problem in Norway. Quantitative research for this period shows an increase in the numbers of available computers and computer knowledgeable teachers, but only minor changes in the relation between boys' and girls' computer interest. We can thus say that the numbers indicate a positive development, but that there still is a problematic relation between girls and computing, especially when compared to boys' interest in computers.

Seduced by numbers?

The narratives cover many stories On the basis of quantitative data on computer use and gender in Norway, we can see that boys have greater access to knowledge making in computer mediated environments. Qualitative research results show, for instance, that the differences are most apparent regarding computer games and technological conceptions. Only a small minority of the boys know the exact technical definitions while many boys are not particularly confident on the definitions (Vestby 1998.) Qualitative research has also shown that many girls do not perceive their own computer use as proper computer use. ‘Proper computer use' is defined as useful computer use like daddies do at work, and ‘proper computer users' are anti social boys who have no friends and spend all their time in front of the computer (Gansmo 1998). Additionally, qualitative research shows that many young girls think it is fun to play with the computer, communicate and explore the different programmes on it, and what happens when they push the different buttons (Håpnes and Rasmussen 1997, Gansmo 1998, Nordli 1998, Kvaløy 1999). When qualitative research shows that some girls' definition of ‘proper computer use' is not in line with the use they make of the computer themselves, we should heed a warning that the numbers are not necessarily so clear and obvious after all, and be suspicious that, in many quantitative surveys, such girls might have reported their own computer use and knowledge as being too low. Qualitative research opens up for differentiated understanding of gender by for instance showing that some girls understand the computer as a fun toy they have full freedom to play with, other girls play with the computer but claim they are using it for useful tasks like information gathering and school work. A third group of girls finds computing dead boring, but since computing is regarded as the key to the future they feel they have to be smart and learn computing for useful purposes. A fourth type of girls finds computing the most exciting activity in the world, and use it just as much for fun as for useful tasks (Gansmo 1998, Nordli 1998). When girls can be so different and make such different use of the computer, it seems natural that they also define ‘proper computing' differently. Qualitative research has shown different levels of computer confidence and knowledge among boys, that girls seem to be intrigued by the communicative aspects of Internet, that girls think computing is fun, and that girls have different interpretations of computing. Thus the qualitative research

187

188

Researching ICTs in Context

results challenge the dichotomous understanding of gender and the notion of a 'gender divide' when it comes to computers and learning.

Use of research results in policy-making and public administration Social science research is increasingly funded due to an expectation that the research results somehow will be directly useful for society as a whole (Naustdalslid & Reitan 1994). The public debate on the usefulness of research rarely separates social science research from other kinds of research. Much of the debate is based on the assumptions we have of how scientific research results may enable us to solve specific problems. The idea is that social science can provide humans with similar control over society to what we eventually have gained over nature (ibid.). Sociology of knowledge has traditionally drawn a distinction between (natural) science and non-science subjects like art and social science. The latter has been regarded as the outcome of its historical and social context, while the laws of (natural) science have been seen as independent of their context. Such science is seen as objective, value-free knowledge which is independent of the observer. This positivistic view has become part of common-sense knowledge of our society. Even though we know little about nature we accept scientific findings about it to be true (McNeill & Townley 1981). On closer examination we may discover that the evaluation of social science's usefulness is based on the same ideal as the natural sciences. Research results should provide decision makers with solutions to problems they otherwise would not have been able to solve (Naustdalslid & Reitan 1994). Research is therefore evaluated more on the basis of its practical usefulness than for its status in academia (Weiss 1978). This makes it likely that decision makers are not really concerned with whether the information is based on true or scientifically consolidated knowledge. Indirectly this might still be important because decisions based on questionable knowledge might be drawn into question and be thrown back at the decision makers. It is therefore likely that decision makers would like to cover their backs by applying only what they perceive as true research results (Naustdalslid & Reitan 1994).

Seduced by numbers?

That natural scientific solutions are often understood as more objective and universally valid might make it easier to a overlook or reject social scientific suggestions. In cases such as that of girls and computing, where more or less only social scientific research is conducted, it is likely that qualitative research solutions are dismissed because the quantitative research solutions are perceived as more naturally scientific and, hence, more true. Power is gained through figures and statistics and those in power know how to stay in charge by using figures and statistics (Lie & Roll-Hansen 2001).

Research results applied in different ways It might be just as relevant to ask why research results are applied at all as investigating why it is not being used. Doing that we can distinguish between four main categories for use of research which are not necessarily mutually exclusive: informative, instrumental, strategic and symbolic (Naustdalslid & Reitan 1994). The informative application of research (un)consciously contributes to shaping our notions of society, our conceptions of cause and effect, and the perceptions of current problems. This kind of research application is to a lesser extent aimed at specific decisions, but more towards having an effect on how society works in general (ibid.). Instrumental use of research, on the other hand, means concrete, goal oriented use of research results to solve specific problems. In order to use knowledge from research results instrumentally, the results must have a possibility to inflict on the social realm (ibid.). Strategic application of research results is conducted when decision makers who are convinced they have the right solution, choose to apply research in order to communicate their opinion and convince others that they are right (Langley 1989). This is based on an idea that arguments funded on research will strengthen their stand and contribute to giving the opinion legitimacy (Feldman & March 1981). This does not mean that the research it self is biased, but that the policy maker applying and presenting it is biased (Nilson 1992). In order for the strategic application of research to be effective and credible, it is crucial that it comes out as instrumental, and as an unbiased expression of truth (Naustdalslid & Reitan 1994).

189

190

Researching ICTs in Context

Symbolic use of research results occurs because the public administration is part of a ritual game where the participants must fight for their credibility (March & Olsen 1989). It might be more important for politicians to make it look like they are active and goal oriented than to make the changes actually happen. Initiatives to fund research for instance on girls and computing might be presented as an interest to actually do something about the problem (Naustdalslid & Reitan 1994). Thus, research might be applied to give a symbolic impression of rationality and efficacy that in turn contributes to the credibility of the decision (Langley 1989), and hence the symbolic application of research is strategic (Naustdalslid & Reitan 1994). In such cases, it is of no importance what the research results say, but I will presume that the more rational the research seems, the better, and, in practice that means scientific research rather than social science, and numbers rather than narratives. This goes along with the observation made by Feldman and March (1981) that intelligent choices and systematic and goal-oriented application of information is perceived as the ideal basis for decisions in western ideology. To possess information in itself gives credit to the organisation (ibid.).

When research is not applied There may be several reasons for not applying available research results in public administration. We may look for the causes among the users, within the research itself or in the interaction between these two worlds. Simplified this might be summarised in three not mutually exclusive reasons for not applying research. 1) There might be organisational issues among the decision makers which keep them from using research. 2) The research is not relevant or is not made easy accessible for the users due to layout, academic quality or methodological approaches. 3) There might be cultural differences between the two worlds causing problems with the 'translation'. Research application is additionally a question about the individual characteristics of the decision maker, their education, trade and beliefs, and a question about structural features of the user organisation (Naustdalslid & Reitan 1994). Good research is not necessarily equivalent to relevant research. Good research is known to study the matter in depth, provide information on relevant issues and details in order to find new and surprising

Seduced by numbers?

results – which will make the decision process more complicated. Relevant research on the other hand is available in a simple and surveyable form when the decision maker needs it, and it must provide knowledge that can contribute to making the decision process easier (ibid.). Research is not particularly relevant if it is mainly concerned with explaining social phenomenon that cannot be influenced through decisions. This might explain why research concerned with big social issues and institutional systems, class and power, does not produce a type of knowledge politicians and policy makers feel they can use to solve specific problems (ibid.) It is likely that much of the qualitative research on girls and computing has provided exactly this kind of explanations, explanations about construction of gender due to institutional relations, and that this is one of the reasons why policy makers find it hard do apply and operationalise qualitative research results on girls and computing. Knowledge gained from qualitative research is not necessarily hard to access, but it can make the decision process more difficult, and it sometimes points at explanatory variables which are perceived to be out of reach for the decision makers, such as the argument that computers are not perceived as a uniform technology, but used and perceived very differently by different people.

Translation of research results into socially robust knowledge Recent research, conducted by studying the knowledge production and application as a process of negotiation between different actors, has contributed with another insight into how research results are (not) applied and interpreted, and how research results are (not) granted status as authorised facts (Latour 1987, Gibbons 1999, Nowotny, Scott & Gibbons 2001). The laboratory as a metaphor for all research is central in all decision making because most controversies mobilise scientific expertise. Research results are used to grant legitimacy to political goals and means (Latour 1987). The research process might be described as a journey that in the end goes from the researcher's publication to the policy maker's interpretation of it, which again leads to either rejection or application in some way. When something is on the move, like for instance research results,

191

192

Researching ICTs in Context

new understandings might occur, for instance due to the network the research results are perceived to be connected to, both regarding research institutions, other results and the experiences of the user. The translation process is central in Latour's approach, and we might say that politicians translate the relevant research into narratives. The interpretation of the research is delegated to the users, and they may choose to interpret it differently than the researcher intended (ibid.). This implies that even though research is mobilised as a resource in controversies, research is not necessarily granted power or authority. Gibbons (1999) notes that there has been a social contract built on trust between society and three relatively independent actors of scientific research: universities have generated fundamental knowledge for society. Industrial research and development departments have carried scientific knowledge into products and innovations. Finally, governmental science has filled the gap between the two other actors. Under this contract, science has been expected to produce reliable knowledge and to make it available to society. Changes in the relation between these three actors (for instance due to the end of the Cold War and a change in the interests of the state), the expansion of higher level education, an increasing focus on objectdriven research and a desire to get something in return from the research-funding, has led to the need for a new social contract. Science is now produced in more open systems of knowledge production. Thus a complete re-thinking of science's relation to the rest of society is needed (Gibbons 1999, Nowotny, Scott & Gibbons 2001). The price of this increased complexity is a pervasive uncertainty, and erosion of society's stable categorisations and of the ideas of truth and authority. Consequently science can no longer communicate its reliable knowledge to society through a monologue, as it has done successfully for the past centuries, but society can now speak back to science. In the old contract the autonomy of science was seldom questioned, and science was seen as basis of all new knowledge which it was expected to communicate to society, which in turn was supposed to transform the knowledge into action. It is less often appreciated that society is also transforming science as it speaks back (Gibbons 1999). Science produces reliable knowledge, not in terms of objectivity, but in terms of what works for the relevant peer group. What works has now acquired a further dimension in a shift from reliable knowledge to

Seduced by numbers?

socially robust knowledge because knowledge must be contextualised and science must listen to what society speaks back. Socially robust knowledge is valid both inside and outside the laboratory. This validity goes further than the relevant peer group and involves an extended group of experts and lay-experts. Since 'society' has participated in the making of this socially robust knowledge, it is less likely to be contested than reliable knowledge (Gibbons 1999, Nowotny, Scott & Gibbons 2001). In the new system, knowledge will no longer be superseded by superior science, but rather it may be sharply contested, and it is no longer within the control of scientific peers. This involves renegotiations and reinterpretations of dramatically extended boundaries. Science can no longer be validated as reliable by conventional discipline-bound norms. In order to become robust, science must be sensitive to a wider range of social implications (Gibbons 1999). One outcome of these changes is that the problems are now formulated and negotiated in the agora – where science meets the public and they speak to each other – rather than in the university, industry or in government. Media is increasingly active in the agora, and here new communication technologies play an important role. The contextualisation occurs in the agora where today's societal and scientific problems are framed and defined, and their solutions are negotiated. Thus, the role of the scientific expertise, which has been so important in decision making, is changing as expertise spreads throughout society and splits up the established links between expertise and decision making/management (ibid.). The authority of science will need to be legitimated again and again, and to maintain this, science must enter the agora and participate fully in the production of socially robust knowledge (Gibbons 1999, Nowotny, Scott & Gibbons 2001).

Methodology The analysis presented in this chapter is mainly based on qualitative interviews with five Norwegian policy makers in the Ministry of Education and Research (UFD) concerned with implementing computing in primary and secondary schools and four politicians representing four different political parties in the UF committee.1 The interviews were conducted during spring and summer 2000. During the interviews we talked about the decision makers' ideas about implementing computers

193

194

Researching ICTs in Context

in secondary schools and the fact that girls have been regarded as not as competent and eager as boys in front of the computer. Additionally we talked about the decision makers' knowledge base on the issue, where they gathered their information and what kind of research they used to form their decisions. Previous to these interviews I had analysed several white papers from the Norwegian government and their programmes on ICT in education for the period from 1983 to 2000.

Application of research to solve the problem 'girls and computing' Organisation

The different ministries review questions and develop suggested solutions on behalf of the government who in turn present these cases for the parliament. Then, the different committees make a proposition, which makes the fundament for the negotiations in the parliament on allocations of funds and new laws. The aim in this case is to allocate enough resources to ensure better access to computers in schools. The general tasks for the ministries and the parliament concern a special focus on allocation of funds. It is thus obvious that the decision makers' need for knowledge is oriented towards numbers they can enter into the equation for distribution. The political information need is first of all covered through information from the ministries. But the politicians can also base their decisions on additional results. They invite special interest groups like the teacher union, groups of parents in the primary and secondary schools and others to voice their opinions. The members of the UF committee have generally little knowledge of the subject since girls and computing has not been treated as an issue. They support their decisions regarding girls and computing on knowledge which they regard as parallel, a point I will return back to later. The ministry is to a larger extent than the politicians expected to posses expert knowledge on girls and computing. The Ministry of Education and Research is organised in different special departments along with an Equal opportunity secretariat that is supposed to provide insight for all departments and make sure every one considers the equal opportunity perspective. Even though the Equal opportunity secretariat

Seduced by numbers?

possesses broad knowledge on what equal opportunity can be and how to implement equal opportunities, it seems like 50/50 distribution of boys and girls is the easiest and most applicable way to understand equal opportunities. In the same way as decision makers find it hard to translate qualitative research into practice, it seems hard for the Equal opportunity secretariat to accomplish anything but a goal of 50/50 distribution within their decision field. As I have shown, the organisation of the parliament and the ministries is directly aimed at applying knowledge and gather information which should be transmitted simply and directly within the ministries and to the decision makers with specific needs. Still, there might be aspects with this organisation that leads to research results being overlooked, or for results to be applied in such a manner that they do not cause direct effects. The topic for this chapter is not the potential organisational problems, but rather the constructions made by the decision makers. I will therefore in the next part examine further the decision makers and their knowledge base.

The decision makers' knowledge base All of the public administrators in this case have some sort of previous connection to the educational system that make their competence interesting to the ministry. They have previously worked as teachers, school managers, in municipal administration of education or with the development of means of education. Some of these have their education within natural sciences, while the consultant in the Equal opportunity secretariat is a sociologist. Among the interviewed politicians in the UF committee one is an economist, two are trained teachers in language and history, and one of the female politicians is trained as project secretary in addition to having a degree in computing from 1989/-90 (when actually quite a few women found computing interesting). The decision makers' background and trade is thus relevant for their work on education in general. In order to handle special issues like girls and computing they have to look to research. The economist in the UF committee prefers to rely on quantitative research result, numbers, which he considers as solid facts. He dismisses qualitative research results, narratives, as hear-say because he finds that the social sciences have not developed reliable methods. The

195

196

Researching ICTs in Context

economist has more confidence in the quantitative research methods due to their system of excluding foolish answers and discovering false answers from the informants. The sociologist in the Equal opportunity secretariat on the other hand claims that it is important that sociology looks into the matter; she mentions several qualitative research findings to prove her point. The other decision makers are to be found somewhere in between these two stands in regard to what is considered important research. Most of them deny relying more on numbers than narratives in their decision making, and that they perceive narratives as less scientific than numbers. However, their choices and interpretation of the relevant research shows another picture.

What do the decision makers read out of the research? I base my decisions on the available research for the current issue. Some of it is old, some of it new. Some of it is very narrow and to the point, and some of the research is very broad. (Politician in the UF committee).

The public administrators apply research systematically in their work and order reports from central research institutions. Knowledge regarding girls and computing is gathered through the Equal opportunity secretariat. In addition, public administrators spend time on the Internet to search for information, they investigate what the universities produce, examine the newspapers, and try to participate in national and international conferences. They consult with colleagues in other countries and read relevant journals. This seems as a thorough and active search for, and effective internal distribution of, relevant research results since the same selection of research reports are continuously mentioned by the policy makers.2 During the interviews the decision makers constantly referred to 'research shows' and '(foreign) experts claim'. This demonstrates the important role research plays in their knowledge base and as authorised facts. Why and to what extent research is applied, and how the research findings are interpreted is still far beyond the control of the researcher. Most people in Norway seem to agree on the importance of computing as a useful tool for everyone. It is thus regarded as a political problem that research shows that not all citizens have access to or know how to use a computer. One of the politicians meant this problem would solve it self eventually, but in the mean time she argued for the

Seduced by numbers?

importance of schools as the great equalisers in making everyone friendly with the computer. Further, computers are regarded as important because the Internet offers easily accessible information, a feature which especially girls are perceived to prefer. Public libraries should therefore have available computers to secure access for everyone. The claim that it is crucial to be familiarised with computing at an early age is seen as the main findings of the research, at the same time as they point to British experts' concern over this early introduction to computing. Computing is regarded as both important and 'dangerous' at the same time. In some aspects computing is seen as dangerous and in others as purely positive for the users. This goes to show the problems that arise from perceiving 'computing' as a uniform and unambiguous technology. Additionally, research which reports that the pupils often are better than their teachers at computing, and that men still dominate the computing industry and higher education causes concern. Research is thus said to show that it is important to focus especially on girls and women because they have been engaged in computing to a lesser extent than men. There is also a widespread notion that girls ‘do not want' to learn computing. An IBM research report from 1999 apparently shows that Norwegian girls spend less time on the Internet and computing in general than their sisters in the other Nordic countries. The fact that Norwegian boys also spent less time than other boys in the region is hardly mentioned, and it does not contribute to adjusting the notion of the reserved girls. It is also a paradox that computing skills and interest is measured by numbers of hours spent in front of the computer. When comparing girls' and boys' computing, the common interpretation of research is that the more hours spent in front of the computer the better. On the other hand, there is widespread fear that the children will grow up as antisocial creatures because they spend hours alone in front of the computer. This fear is totally absent when decision makers express concern due to research reports showing that girls spend less time (= are less interested and competent) than boys in front of the computer. It is also a paradox that no one voices the alternative interpretation that girls might be better than boys at computing and therefore do not have to spend hours in front of the computer in order to finish their tasks. Additionally, when qualitative research results on girls and computing show how different the girls are from each other, this is often interpreted according to the quantitative notion of two op-

197

198

Researching ICTs in Context

posite genders, and girls are thus seen to be better at the social and communicative parts of computing than are boys. The decision makers claim that in comparison with other countries they find large differences between Norwegian girls' and boys' competence in natural science subjects. Most decision makers think this also applies to computing. They also read into the research reports that girls are communicative while boys approach the technology differently. Communicative use of technology is implicitly perceived as subordinate when the goal is perceived as 50/50 distribution of girls and boys in higher computing education. It is also possible that different applications of computers are seen as almost irrelevant, as long as girls are outnumbered in front of the computers and in hours spent computing. My assertion is that this narrow but seemingly all-embracing and universal understanding of the term 'computing' that is seen in almost all interpretation of research reports and in the decision makers' knowledge base, stops them from seeing alternative notions of what computing skills could have been in primary and secondary schools and how this is best implemented for everyone. Additionally the difference between boys and girls are stressed in these interpretations rather than focusing on the narratives which emphasise the diverse gender and the differences between boys and between girls. It seems as the qualitative findings are overlooked and all power is granted to the quantitative research on the problematic girls and computing relation.

Numbers are easier to remember Most of the decision makers are uncertain whether the figures are any better in other countries. This goes to show that figures reveal and pinpoint problems: I don't know if girls and computing is a problem in the rest of Europe as in Norway because I haven't seen any research statistics on that. (Politician in the UF committee).

Further, the number of female teachers in primary and secondary schools causes concern among the decision makers. This is because these women lack computing skills and interest, and because male teachers are more or less absent on this level of education. The male dominance in the computing industry and education are other numbers

Seduced by numbers?

that give the decision makers reason to worry along with the lack of sufficient computing equipment in schools. Polarisations and presentations of differences between few and simple categories seem most interesting and easiest to read out of the research reports. Gender is thus a natural category to focus on because gender is perceived relevant in connections with computing and because gender is seen as easy to 'measure'. Policy making is described as a mixture of exactness and trends, and it seems fit to allegate that the decision makers perceive numbers as representing the exact while narratives represent trends, and thus are granted less authority because they are seen as less scientific. If qualitative research is based on other research on facts, that is not to be despised. A lot of research can be conducted based on previous research on facts. You can for instance conduct a relatively narrow survey, and apply the broad material together with your own narrow population and make other issues and conclusions. So I don't see any reason why everyone should do research on everything all the time. We do after all have good access to facts that can be applied in many aspects. (Politician in the UF committee).

Decision makers claim they base their decisions on research reports because good research is their best aid and because research is said to provide answers for the future. In the case of girls and computing there is seemingly no controversies which call for the need of scientific expertise to settle arguments with authorised facts. The research seems to produce and make available knowledge on youths and computing that legitimises the allocation of more funds to computers in schools, and thus continues to stress the difference between boys and girls and the idea of a necessary 1:1 relation between pupil and computer in order to learn anything. Despite claiming that numbers and narratives are of equal value, many of the decision makers admit that quantitative research results are better because they need to apply figures in order to focus on how computing can be implemented for everyone. In one way you might say it is easier to make generalisations based on the figures. When you are to make a political decision you ask if there has been any change, can you say anything about the figures. That can be difficult with the qualitative research results. You may lean on positive effects in this and that school, but usually you need the figures as support as well. If the one research supports the other, qualitative research is fine.

199

200

Researching ICTs in Context

We need both, but we often see that for instance gender studies apply qualitative methods, that is smaller research, and it is not easy to make generalisations from those to go for the entire society. Even though they supply important knowledge, we also need the quantitative research results. The research conducted by SSB is very important. Everyone likes it when, here we have conducted a national survey where all schools have participated, and where we can see that 27% this and 60% that. In order for a need to be specified and treated we need its numbers. (Public administrator in UFD).

In addition, policy makers often dismiss qualitative findings as hear-say anyone can make up on their own. If this 'hear-say' is in line with the policy makers' own experiences they are inclined to use the qualitative research results as support for their own opinions in discussions with others. If not, qualitative research is brushed off because of its lack of connection to facts (that is: numbers). Qualitative research is perceived as small scale research. Further, most decision makers constantly stress the weaknesses of qualitative research, such as the narrow choice of population, a wrong problem focus and lack of generalisation effects. No one raises a single, critical question towards quantitative research nor the interpretations and applications of numbers. Several of the decision makers tried to give me examples of relevant qualitative research results that obviously are quantitative.3 Some of the decision makers claim that they actually apply narratives because girls and computing is seen as difficult to describe in numbers. And they also point to the fact that the numbers are based on assessments that apparently make the research more qualitative, or put in other words, maybe of higher quality? Research on the competence of teachers and available teaching aids is said to be just as important as research on numbers of computers in schools. The previous is described as qualitative features and research on the topic is therefore considered qualitative. Overall, there seems to be little room for the actual qualitative research results.

Big brother and little sister as competitors for attention The decision makers have a hard time separating properly between qualitative and quantitative research. Decision makers and often the general public as well construct quantitative research results as more scientific and hence more true and based on facts. Qualitative research on the other hand is perceived as focusing on the quality of a phenome-

Seduced by numbers?

non, predicting trends or as the little sister of research on facts. The little sister is younger and not as wise as the big brother and has less capacity to take on large tasks, and hence she is granted less power and auhority. The decision makers are not researchers and, therefore, can not be expected to know the differences between qualitative and quantitative research methods. None-the-less, their construction of quantitative research as big brother and qualitative research as little sister has fatal consequences for their choice of knowledge base. In this regard, it is also interesting to note that the decision makers refer to quantitative research results using the names of the powerful institutions like SSB and IBM, while qualitative research results often are referred to by using only the first name of the specific researcher, like Dagny, Bente and Tove. Numbers might thus be granted more credibility because it seems as an entire research institution has produced the research while the narratives are too tightly connected to one, usually female researcher. Since, a limited extent, the decision makers to know the differences between qualitative and quantitative research (but still construct quantitative as 'more scientific'), the numbers of researchers perceived to be responsible for the research may be a determining factor when it comes to reliability. Further, there seems to be an unfortunate connotative link between the sex of the researcher, critical research conducted on gender and the preferred social scientific methods applied - that most of the qualitative researchers on gender issues literally are little sisters who are not conducting 'real' science since they often apply qualitative methods. As seen in the quote from the public administrator above, '…we often see that for instance gender studies apply qualitative methods, that is smaller research…' that can not compete with 'real' science of big brother.

Personal experiences as knowledge base Information on new phenomenon and problems must be delivered through numbers. Without such 'facts', decision makers refuse to believe, for instance, that female-related or -generated websites on the Internet have become a success. But it is likely that the decision makers would have talked about the successful female websites if their own daughters had told them that they regularly visit such sites when surfing the Internet. Personal experiences and experiences of family and friends

201

202

Researching ICTs in Context

seem to be just as valuable as research results in the decision makers' knowledge base. Some decision makers admit lacking information on girls and computing. Especially in the beginning they had to rely a lot on personal experiences in order to focus on how to implement computing in order to take girls' best interests into consideration. Officially, to a large extent the decision makers base their decisions on relevant research. Still, just as often as they mention research results, personal experiences or observations of their own children in front of the computer is mentioned.4 Several of the decision makers told me that their daughters were instrumental towards computing and that they showed little interest in the computer, in accordance with the traditional perception of girls and computing. Some of the daughters were considered as interested in computers due to a fascination in computer games like Pippi Longstocking. One of the decision makers told me that his two daughters and one son had altered his view of girls and computing. In his family the girls were more interested in computers and had greater competence in using them than the boy. Another of the decision makers experienced that technology oriented families served as inspiration for girls. Yet another elaborated on the observations she had made during visits to good friends in a poor European country. She was very impressed with what can be achieved with little money, and concluded that lack of money opens up for creative implementing of computers. Several of the decision makers applied their experiences from their studies or from their workplace as a knowledge base for considerations and decisions pertaining to girls and computing. It is interesting to note that most of these experiences hold women as the most eager and competent computer users while men are seen to be less competent and more afraid of asking for assistance. Personal experiences and observations thus play a considerable role in the decision maker's knowledge base. Isolated and casual 'qualitative research' conducted by the inexperienced 'decision maker researcher' seem to qualify as important arguments in the debate on girls and computing, while qualified narratives which to a large extent show the same tendencies as the aggregated casual narratives, are diminished due to lack of universal value and reliability. It is a paradox that the qualified and casual qualitative research results show the same tendencies: when it comes to computers, girls use and comprehend computers differently,

Seduced by numbers?

and many of them are indeed clever in understanding them and fascinated by their use and potential. Still, the official perception and the attempts to solve the problem are aimed at rescuing the scared outnumbered girls in order to reach 50/50 distribution in front of the computer.

Parallel fallacy The UF committee has not treated girls and computing as a special issue. UFD is concerned with implementation of computer into schools in general. This might explain why the decision makers to such an extent need to rely on personal experiences rather than research. At the same time, they consider research results as an important base for decisions since they constantly refer to research from a field perceived as parallel to, or maybe even as the same problem – girls' problematic relation to natural science subjects. Compared to other countries there are large differences between boys' and girls' knowledge of natural science. Most decision makers think the numbers are about the same for computing. One of the decision maker's daughter has a Ph.D. in natural science, and she has had a hard time fighting her way. Based on her experiences, her father concludes that girls are afraid that computing is related to natural science, and thus choose not to study computing. The option to do something about this alleged misconception is not mentioned by any one of the decision makers. As shown previously the decision makers rely heavily on numbers. For instance, they do apply quantitative research on attitudes where the numbers show that girls are more negative towards mathematics than boys are. The numbers show that girls are more or less absent, both in computing and mathematics. The decision makers draw from this that the same reasons are causing girls to be outnumbered - girls' fear. The problem in focus might have been another one if we had measured something besides the numbers of girls versus the numbers of boys in the different subjects, and we could have avoided what I call a parallel fallacy. Numbers that show the same differences seem to mislead the decision makers to believe the same causes are to blame. At the same time, their own experiences and notions of what is perceived as difficult also may lead to parallel fallacies. Several of the decision makers have experienced that computing as tool for office work is easy to learn. Programming on the other hand is perceived as very difficult.

203

204

Researching ICTs in Context

They do not understand how it works and think that in order to become a computing engineer you need a lot more knowledge than they possess. Simultaneously, few of the decision makers have their degrees in natural sciences, and perceive natural science as very complicated. It is likely that all areas perceived as complicated are brought together and constructed as one arena that can be treated with the same means. In addition to treating girls and computing parallel to girls and natural sciences, the decision makers rely on general knowledge on traditional gender roles found in different reports. They assume that the same traditional division also applies for computing. Their perception of traditional divisions imply that girls are scared and afraid of trying new things, and that boys are nasty bullies who force the girls away from the computers. The repeated countings that presumably show that girls are outnumbered are interpreted in terms that construct girls as passive victims who do not see what would be in their own best interest. In several contexts where girls are outnumbered, for instance in top positions in private enterprises, one of the reasons might be suppression and exclusion of girls due to male dominance which excludes girls from interesting positions. We can not automatically draw the conclusion that the same suppressers and strategies of exclusion are at work when the numbers show the same tendencies in front of the computers in primary and secondary schools. Quantitative research has detected unwanted differences between boys and girls through repeated surveys. Public statistics have always played an important role in public decision making. Numerical information is continuously becoming an important resource in decision-making and is a means to be granted more attention (Lie & Roll-Hansen 2001). Such facts are perceived as unnegotiable, and no one seems to be critical towards the origin of the numbers or what they actually can tell us. When numbers are given such a large role in setting the agenda, research might actually work to conceal facts. The decision makers continue to base their decisions on numbers that describe girls as one homogenous group in contrast to a corresponding group of boys, despite all the qualitative research reports which show that the category 'girl' includes so many different and heterogeneous human beings who should not be placed in the same box and that computers are used and interpreted very differently. Thus this contributes to concealing other differences which might have been relevant in order to solve the problem.

Seduced by numbers?

Large potential for better application of research on the problematic girls and comput computing relation One of the decision makers found it very hard to answer whether and what type of research she applied in her work, because she found so much research to be biased. She claimed that such biased research would not go far within the policy making, but she found it hard to rely on research when the results often are contradicted by other research findings, or when the researcher only has focused on one aspect of the matter in question. In addition, she claimed that many apply research like the Devil reads the Bible or only apply research when it supports their stand. In a perfect world, all good research would have been applied instrumentally to solve problems, and to inform and make decision makers and society in general more aware of the issue, and society would speak back to science. The political system and the public administration is far from part of a perfect world and much more likely to be parts in a strategic game: I base my decisions on The Curriculum, some on research, some on the possibility of being part of a majority in order to actually do something. There are several aspects. (Politician in the UF committee).

One of the politicians in the committee admits that personal experiences are crucial for the choices she makes of reliable research: The research I've read is mostly conclusions on... there are a lot of numbers there... but the conclusions, the summaries, more or less go along with my own experiences. And that is maybe why I choose to apply it as well (smiling). There is a tendency that we find research that supports our experiences and beliefs. (Politician in the UF committee).

One of many aims for the actors in this political and administrative game is to implement computing into secondary schools in a way that makes girls interested, and thus full equal opportunity can be achieved. They are searching for the good and effective decisions to solve the problem with absent girls. Since social science, and especially qualitative social science, research is perceived to be scientifically questionable by many, decision makers are more likely to make use of quantitative research results which are perceived as more natural scientific and hence more true. Qualitative research results that claim girls do not act as one homogenous group in front of the com-

205

206

Researching ICTs in Context

puter might also be easier to dismiss because they make the process of decisions more complicated. To a certain extent it might seem as the decision makers are too quick in their decision making, relying on insufficient, or what they perceive as parallel, information. Some of the decision makers do apply research results instrumentally, but mainly numbers, which have not been especially suitable for solving the girls and computing problem since it is still regarded as a problem 20 years after being put on the agenda. In some cases, the decision makers apply research results strategically to prove their point. To a certain extent, some of the research is only applied symbolically to convey a notion that they are aware of the problem and are trying to solve it. Many of the numbers contribute to putting items on the agenda and thus function as a concealer for other problematic areas. Despite the organisation of the public administration and the parliament in order to gather information from relevant research and apply this in their decisions, I have shown that choices of information have a certain bias. To a certain extent some of the decision makers claim to reject research because they find it not to be of high enough quality or that it is conducted wrongfully. These are important aspects of quality, but as shown previously, the decision makers are not trained to make such evaluations and thus make mistakes, or, as a paradox, conduct their own casual qualitative research. One of the main reasons for overlooking relevant (qualitative) research results on girls and computing is likely to lie in the cultural differences between the users and the researchers. The user translation of qualitative research results makes it appear as little scientific. This is an important aspect the qualitative researchers need to consider in their future work, for instance by making more generalised conclusions, disseminating their findings in a more interesting and comprehensible way, and positioning themselves together with large and well-known research institutes, in order to gain more reliability. Quantitative researchers, on the other hand, should make efforts to disseminate their analysis of the numbers rather than the numbers themselves, in order to guide the decision makers to professional interpretations of what the differences between boys and girls indicate and what might cause the differences. At the same time, policy decision makers are facing important challenges to become more aware of the weaknesses of quantitative research, to become more sceptical of the numbers, and to learn to understand the numbers and place them into context. They should also become more

Seduced by numbers?

aware of the advantages of qualitative research. But the world is not that simple, anymore. Controversies are seen to mobilise scientific expertise, but in the case of girls and computing there is seemingly an idyllic lack of controversies. This opens up for more individual interpretations by the decision makers, and they can apply research on the differences between boys and girls to grant legitimacy to political goals and the means to allocate more funding to implement more computers into schools in order to achieve 50/50 distribution of boys and girls in front of the computer. Science aims to produce reliable knowledge and to communicate this to society, in this case mainly represented by the decision makers. The decision makers are delegated the task of translating the research results into action, and, thus, they have opportunity of whether or not to grant authority to the research. In their interpretations, narratives and qualitative research results are not taken into account and interpreted as reliable knowledge. Or, the main findings from the qualitative research — in this case, the fact that girls are heterogeneous in their use of computers as well as boys are, and thus should not be categorised in two opposite categories — are interpreted according to the traditional quantitative gender categories. This results in the binary construction that girls prefer the communicative aspects of computers while boys are more inclined to play computer games or investigate the technology. According to the new social contract between science and society, science is produced in a more open system which should contribute to making the reliable knowledge also socially robust. In the case of the translation of qualitative versus quantitative research results on girls and computing, qualitative research is produced openly, inviting the user to validate and interpret the research, while the quantitative research is based on calculations and negotiations that are not usually open or available to the decision maker. The open process of the qualitative research process leaves it open for critique, while the final numbers from the quantitative research, paradoxically, are interpreted as authorised facts. Society's ability to speak back to science and the erosion of stable categorisations and ideas of truth and authority should open possibilities for the qualitative research results on ambiguous understanding of gender and computers to become socially robust. Rather it seems like the complexity this opens up for lead the decision makers (and society in general) to search for order and simple systems of categorisations that stress the dif-

207

208

Researching ICTs in Context

ferences between the genders. Narratives are thus not constructed as socially robust knowledge because numbers are presented more systematically and as plain facts in a simple and easily comprehensible narrative. In order to develop socially robust knowledge, science must also be sensitive to a wider range of social implications. This includes for instance the personal experiences of the decision makers and the system of information translation they must operate within. UFD need to accumulate and disperse knowledge. There is of course a limit to how much information that can be spread and accumulated, and thus we may observe a preference for numbers over narratives. But this has more to do with the presentation of the research results than the method behind the knowledge production. The same need for easy accessible and surveyable facts are also observed in the media and in the agora at large. I have shown a variety of multiple explanations on how and why (not) research results are applied in policymaking and what kind of research is perceived to be reliable and trustworthy. What we clearly see is that decision makers seem to be seduced by numbers since numbers are seen as more reliable, socially robust or applicable than narratives. What we must avoid in the future is the narrow and stable categorisation of numbers versus narratives — just as girls are perceived in opposition to boys — but rather open up for varieties of gender and narratives as the new social contract develops.

References Feldman, M.S. & March, J.C. 1981. 'Information in organizations as signal and symbol'. Administrative Science Quarterly, 26. 171-186. Gansmo, H. 1998. Det forvrengte dataspeilet. En kvalitativ studie av hvordan ungdomsskolejenter forstår datateknologiens muligheter i dag og i fremtiden. Trondheim: STS-rapport nr. 36. Gibbons, M. 1999. 'Science's new social contract with society', Nature, Vol. 402. c81-c84. Håpnes, T. & Rasmussen, B. 1997. Internett . jentenett? Ungdomsskolejenters databruk og datainteresser. Trondheim: Senter for kvinneforskning. Skriftserie 7/97. Kvaløy, K. 1999. Fortellinger om moderne, flinke, lekne unge jenters forhold til datateknologi. En kvalitativ studie av datateknologiens rolle i unge jenters dannelse av kjønnsidentitet. Trondheim: Senter for kvinneforskning. Skriftserie 3/99.

Seduced by numbers?

Langley, A. 1989. 'In search of rationality: the purposes behind the use of formal analysis in organisations'. Administrative Science Quarterly, 334. 598-631. Latour, B. 1987. Science in Action. Cambridge MA: Harvard University Press. Lie, E. & Roll-Hansen, H. 2001. Faktisk talt. Oslo: Universitetsforlaget. March, J. G. &. Olsen, J.P. 1989. Rediscovering Institutions. The Organisational Basis of Politics. New York: The Free Press. Naustdalslid, J. & Reitan, M. 1994. Kunnskap og styring. Om forskningens rolle i politikk og forvaltning. Oslo: TANO, NIBR. McNeill, P. & Townley, C. 1981. Fundamentals of Sociology. Cheltenham: Stanley Thornes Publishers. Nilson, K. 1992. Policy, Interest and Power. Studies in Strategies of Research Utilisation. Meddelanden från Sosialhõgskolan 1992: 1. Lunds Universitet. Nordli, H. 1998. Fra Spice Girls til Cyber Girls? En kvalitativ studie av datafascinerte jenter i ungdomsskolen. Trondheim: STS-rapport nr. 35. Nowotny, H., Scott, P. & Gibbons, M. 2001. Re-Thinking Science. Knowledge and the Public in an Age of Uncertainty. Cambridge: Polity Press. Vestby, G. M. 1998. Jentene, guttene og IT-begrepene. En undersøkelse av ungdoms forståelse av informasjonsteknologi. Oslo: NIBR. Weiss, C.H. 1978. 'Improving the linkage between social research and social policy'. In Lynn, L.E. (ed.) Knowledge and Policy: The Uncertain Connection. Study Project on Social Research and Development. Vol. 5. Washington D.C.: National Academy of Science. 23-82.

Other Stortingsmelding 24 (1993-94). Om informasjonsteknologi i utdanningen. IT i norsk utdanning. Plan for 1996-99. IKT i norsk utdanning. Plan for 2000-03.

Notes 1

At the time of the interviews in spring and summer 2000, UFD was in the Ministry of Church, Education and Research. After the election of fall 2001, the structure of the ministries was changed, and KUF was transformed into UFD, Ministry of Education and Research. 2 The most important research reports seems to be Guri Mette Vestby (quantified qualitative research on youth's understanding of ICT-concepts), Central Bureau of Statistics – SSB – 1995 (traditional quantitative survey on numbers

209

210

Researching ICTs in Context

of computers per school, number of hours spent in front of the computer per week for boys/girls etc.), Tove Håpnes og Bente Rasmussen 1997 (qualitative research on secondary school girls' use and understanding of ICT. This report is based on Gansmo (1998), Nordli (1998) and Kvaløy (1999), IBM (1999: survey of distribution of computers and computer use in the Nordic countries), and SITES (repeated surveys of equipment and use in different countries). In addition, ITU at the University of Oslo is appointed main distributer of research on ICT in education. 3 For instance Berit Skog's Analogue and digital teaching aids– pupils and teachers in the information society. 4 This is information the informants shared unsolicited with me. In advance I had not defined experiences of family members and friends as a source of knowledge/information. By asking open-ended questions and inviting the informants to share as much as possible of their experiences with me, I discovered this rather interesting paradox: that qualitative research results are dismissed in favour of personal experiences.

10 Researching performance, performing research: dance, multimedia and learning Synne Skjulstad, Andrew Morrison & Albertine Aaberge, InterMe InterMedia, University of Oslo

Figure 1: Willson Phiri & Stephanie Sund in a digitally enhanced duet; video still from performance at Statens balletthøgskole, Oslo, December 2001.

211

212

Researching ICTs in Context

A sense of context One day at a time

It's the middle of the week, early in the afternoon and we are about midway through a collaborative educational and research project called Ballectro.1 As the name suggests, the project is a hybrid of dance and digital media. The slate grey autumn sky bounces up from the Oslo fjord into the large curved windows of one of the studios at Statens balletthøgskole.2 Six final year modern dance students are in a class with their choreography teacher and are joined by three project participants from InterMedia at the University of Oslo. We are workshopping material which may become a part of a dance and multimedia performance scheduled for the end of the semester at both our institutions. We are also collaborative learners in a project investigating dance and multimedia. Yet, as multimedia researchers we are also learning how to carry out research into digital media as performance. Improvisation is central to today's session as is often the case in our ongoing collaboration both as performance and as research. The choreographer has asked the students to select one figure each from a large cluster of postcards and to use it as a springboard to developing a solo piece. The pieces will then be combined in a larger sequence which, later, might be included in the overall performance. The students improvise their solos and then, after a break, they repeat and refine them, this time accompanied by music. This music had been developed especially for the project by a different student who has volunteered his talents to the project. The students develop their movements, some of them very actively, others through more intimate expressions. In the corner, the digital music maker lies fast asleep, stretched out between his studies, a part-time job and composing the music for the project. The choreographer darts between the students, quietly moving them into slightly different positions, whispering suggestions to two of them. She places each card on the floor at the front of the studio, mimicking the spatial zones in which each of the students is moving. The electronic music, still in an early version, creates the sense of future performance, one to be remixed, and reperformed as our collage-like approach becomes more coherent. 'Come over here and take part,' the choreographer says to the two multimedia developers sitting in the corner. 'Just dance,' she encourages with a smile. And slowly we begin to fumble around between the styled improvisation of the students. They seem quite unperturbed by our

Researching performance, performing research

wobbly steps. Later, as the media players in this unfolding scene, we comment to one another that this is what ethnographers do, they become part of the fabric of the context they a researching, if not its soloists!

Background to Ballectro

Collaborative project design

In this chapter we present material from this collaborative educational project involving improvisation, dance and multimedia. This varied for the different partners in the collaboration. In this sense the chapter is about 'learning performance'. We therefore also explicitly used the project as a research agenda generating process. Recent research into the potential intersections of dance and digital media has often carried out through collaborative processes and workshops. For example, Birringer (1996) details the heuristics of one such workshop as being '… offered to provide a laboratory for the organic intergration of performance and digital arts, and for the development of new interdisciplinary methods of composition.' As media researchers and teachers in the field of digital media, we discuss ways of conducting and understanding research as and through performance. We suggest, by close reference to the development and performance of a multimedia-enhanced dance piece, that researching ICTs might be expanded to include notions of performance. We suggest that this is potentially fruitful at several levels. First, that research into ICTs sees context as a significant element in how research is performed. Here, we refer specifically to how to learn to build a new media-enriched performance environment in which the medium of expression is primarily moving bodies and moving media. Second, we suggest that research into ICTs might benefit from considering aspects of performance studies in studying and interpreting digital media and their various representations and uses. Third, through the inclusion of visual material, from both the process of making the dance piece and from its performance, we offer a glimpse of how research into ICTs may to take up some of the challenges of applying of digital media in presenting research.

A learning design Ballectro was based on the collaborative development of a student performance for a non-fee-paying audience. This context gave us freedom

213

214

Researching ICTs in Context

to workshop our way into making an actual performance as well as to be open to ways of researching an interdisciplinary interplay between media and dance, between learning and performance, and between research processes and products. This project design allowed the students to participate at different levels of the project as learners of dance, learners of choreography and, to a limited extent, as learners of new media.3 We were also learners of project-based research involving new partners. As researchers and designers of digital media, our main objective in the following collaboration process was to develop closer understanding of digital media and how dance could inform both production and the research into ICTs. Ballectro was an experimental new media arts project, and by definition it involved a range of boundary crossings. In a discussion of such emergent and experimental work, Stocker and Schöpf (1999: 14) comment that: Even though the only aspect that the heterogenous, hybrid configurations of current works often have in common is their use of the computer – that is, their technological or material medium – an essential, defining feature of this new art is impossible to overlook: despite the experience that has been gained and the virtousity that has developed, media art is, above all, an experiment – one that often brings the creators and the proponents of this new art into association with engineers and researchers.

Many interdisciplinary researchers find themselves crossing fields, methods and theories. They need to also find ways to work within these intersections and find ways to communicate them as research.

A publication design Dance may be an important player in the building of the interdisicplinary links between practice and theory which are needed in the study of ICTs in culture and context. For example, in referring to a dance and digital media workshop, Birringer (1996) comments that: Performance — dancing with and across patterns — is an avenue to contest rigid or vapid formulations handed down to us since the emergence of the strength of bodily intelligence unravels the grids and pixellated monotonies of the computer's inscriptive power.

We take up this challenge by investigating this claim also with reference to the building of a Ballectro project research website in which a

Researching performance, performing research

range of media, including video, are used. In contrast, this print and written text is one formalised means of reporting on a multimedia performance project and its processes of production and public mediation. However, as we argue, the web site provides a more fully mediated possibility for understanding the project. This context refers to the mediation of the content of Ballectro, that is as dance, as multimedia and as research. Yet, it also refers to a way of communicating research centred on a process-based exploratory design. As developers, teachers and researchers of new media ourselves, we have found it difficult to locate an elaborated learning and research context within which to place a project such as the one we present here. Huge financial and technical resources have been invested in ICTs, but perhaps a time is now arriving when we ought to be more seriously considering how the environments, content and contexts in which our digitally mediated communication occurs do indeed intersect.

A multi-method approach The Ballectro project employed a multi-method design in both development and research. In the project we deliberately played with digital tools and technologies — in effect danced with them ourselves. The final performance piece was a collage of elements from a variety of learning tasks, improvisation sessions and more formal plans. Heuristically, the project generated problems which, in turn, needed immediate and longer term research frames, solutions and explanations. Here the spiral model of software development (e.g. Denning & Dargan 1996) and a reiterative and reflexive approach as deployed in the many sub-fields of design studies was important. In addition, production based research was realised through a process of making. These various aspects were extended from researching the process and performance to communicating the project in digital arenas. As with many classroom based research projects, Ballectro was carried out within a broad, but non-deterministic, action research framework. As a choreographer and media researchers, we were instrumental in introducing new elements within the dance curriculum and at a new media and learning research centre. We were actively involved in the production and the research process and had an integrative role here. Further, the research was cyclical: it involved an ongoing interplay of theory and practice (Avison 1997: 198).

215

216

Researching ICTs in Context

The experimental and process character of the Ballectro project as a learning and performance activity, also extends to finding ways to perform research on and in digital communication. We therefore discuss questions about the changing rhetorical and mediational contexts of presenting, reporting and analysing research. This points to the importance of understanding multimedia project design, the application of multi-method inquiry in research processes and designs, as well as in the analysis and presentation of research. In terms of performing research, together these elements may be said to consitute a mutable research design. This is design is flexible, relational, selective, contingent, reflexive and hybrid. Later in the chapter we will discuss such associations and their value in a collaborative design and performance. In the next section we outline some of the difficulties in researching a multimedia-dance process and its performance as text.

Researching performance On performance

While the performing arts have existed for centuries in theatre, dance, and song, it was in the 1970s that performance as a feature of the avant garde become accepted as a medium of artistic expression in its own right (Goldberg 2001: 7). However, the term performance has itself been hotly contested in the complex of post-structualist approaches to media and the arts. Performance is a cultural practice, a pratice of representation, and so inevitably enters the arena of ideology as it often did in the 1970s in the form of happenings and agitprop pieces (Counsell and Wolf 2001: 31). A performance act may question existing systems of thought, actions, and beliefs. It may be characterised as a one-time happening realised through one or several performers. Or, increasingly, it has come to be seen as a more mutable event-oriented expressive discourse. A performance is the text constructed by actors, dancers or narrators, yet its uptake lies with the audience or the computer user. Howell (1999: 146) argues that: As performers you are looking for an 'action language': one you can spontaneously 'speak'…So you need to think by performing, instead of trying to complete your thinking prior to the performance…. Performing

Researching performance, performing research

is not a translation from another language. As in writing, where the hand thinks the sentences, your actions must think the performance.

In recent years many researchers and educators have made claims about the need to develop multimodal, multiliteracies (e.g. Kress 1998, Kress 1999, Tyner 1998). Yet, in their often wide-ranging discussions of digital literacies, few such writers refer directly to dance and the role of performing arts more generally, preferring to remain with broad issues of access, competence and intertextuality. Further, in the now large literature on 'new' media, 'digital dance' has not featured prominently, in comparison, for example, with hypernarrative or computer games. Researchers of games and 'interactivity' (e.g. Aarseth 2002) and writers on the usability of websites (Nielsen & Tahir 2001) are also investigating questions of performance in what Brenda Laurel (1993) has called 'Computers as Theatre'. While concerns such as the user's discourse (Liestøl 2002 forthcoming) in games and the functionality of websites do indeed merit research, the performing arts in digital environments are often overshadowed by more commercially intended products. Further, electronic arts in general have not featured strongly in the research discourses on learning, ICTs and context. In contrast, electronic performing arts are often conceptual and of an embodied nature; they occur both a physical performance texts by dancers, yet they are also constructed by audiences. These performance pieces may occur only online, but they frequently exist in real time, and in material spaces, such as installation pieces in art galleries or as mixed media performance works in front of an assembled audience.4 Dance may also be implicit in these works (e.g. Schiphorst 1996/1997, 1997). Such mixed media performance art is often difficult to categorise. Typically such experimental works play on notions of hybridity, identity, human-machine relations and the subjectivity of the viewer or interactant (Jones & Stephenson 1999). These works tend to be reflexive in character. They draw our attention to the ways in which digital media may be used to mediate performance, whether art, dance or drama. In contrast to computer games, which by definition must be played on screen, or online, many performance pieces which use digital media do so by shifting between different types of screens, projections and media types; together these are part of an emerging live event in which plot, movement, music, and scenography may all be in flux. The intention of the artist is thus often that of mediating this flux to an audience.

217

218

Researching ICTs in Context

The use of digital media in performance now also implies that its elements and structures may also be not only machine mediated but also machine generated. Thus, since the mid-1990s, the contexts of performance have increasingly included elements of digital media in which animation, projection and random selection have entered into the performative text. They have challenged our notions of performance at both the textual and the interpretive level, and further in the intersection, cross-over and hybridisation of these levels. In the print collection Performing the Body/Performing the Text, edited by Jones and Stephenson (1999), a performance script by Coco Fusco and Nao Bustamante is interleafed with more traditional expository academic discourse on performance and interpretation. When multimedia material is included in both performance pieces and in research about them relationships between text and interpretation, 'actors' and audiences may become more complex.

Changing research literacies Research into the performing arts has a long history, one in which critical interpretation has been paramount. The processes of dramatic production or musical composition, for example, have gradually become more central to our teaching and researching the performing arts. This can be seen in a leading British project into the role of practice in research. Hosted at the University of Bristol (Department of Drama: Theatre, Film, Television) 'Practice as Research in Performance (PARIP)' is a five year project which is investigating the intersections between creativity, performance and the broading of research paradigms to include new modes of performance. Such information is to be found online. It points to the emergence of attempts to meet the challenges suggested by both experiments with digital media and performance, and in attempts to convey this interplay between theory and practice via digital media.5 For developers, teachers and researchers of digital media who work with performance arts, an initiative such as this one suggests that there is growing professional concern to establish ways of better understanding relationshsips between ICTs as compositional and mediational tools and the contexts in which they are made, used and interpreted. However, in parts of the academy there is both a lack of interest in these relationships as well as a resistance to understand and to investigate them as arenas for serious research. While there may be research on practitoner-researchers (e.g. Jarvis

Researching performance, performing research

1999) referring to relationships between work-based learning, adult education and practice-based research in knowledge-making, many universities are in the process of learning how to apply digital media in online teaching programmes. Less prominent is a concern with mediating research based on knowledge, experience and insights from new media production and the evolving compositional processes of multimedia research discourse. This points back to the ealier reference to digital or electronic literacies. In the context of such relationships between production and analysis of digital media texts and contexts, these electronic literacies are defined not so much by their realisation via a generation of students who have grown up with digital media, but as part of a potentially modulated research and pedagogical expertise to which the academy ought to already be investing. Many university websites still make limited use of a variety digital media and related research and pedagogical designs which truly stretch new media communicatively, that is in relation to both content and tools. Further, few of the university websites which are concerned with dance could be said to be media-rich. In this respect, a recent symposium hosted by PARIP (which we did not attend) was held to realise its general aim of investigating 'creative-academic issues raised by practice as research, where performance is understood as theatre, dance, film, video and television.' To this we would add digital and online media. The symposium covered four main themes which we paraphrase here as they are similar to issues we address in relation to Ballectro. The PARIP symposium discussed: what practice is as research; how practice may be questioned as research in 'live' and 'recording' media; ways of documenting and re-presenting practice as research; and, the ways in which academic contexts of practice as research affect how it is pursued and evaluated.6 On the research front, many of the online publications about digital media and dance tend to have been written from a dance, rather than a media, perspective (e.g. Wechsler, no date). While this chapter approaches the collaborative Ballectro project from the point of vew of digital media, we believe that it is important that performance and dance be more fullly included in the fast growing field of digital media studies. We argue this because performance arts and performance sudies can contribute to the methodologies for making and analysing digital media. Thet may also play a part in hwo we increasingly come to be users of and actant upon digital media texts, environments and communication.

219

220

Researching ICTs in Context

Performance-based research Drawing on our theoretical and practical knowledge of digital media, design, learning and research, we suggest that we conceptualise performing arts and research about them in digitally mediated domains as performance-based research. In such a research approach, digital media may be part of the design and shaping of the of dramaturgy and scenography. In this chapter, we attempt to support this claim by referring to Ballectro as a pilot in which we were able to experiment with the intersection between practice, production and performance, and ways of documenting, mediating, presenting and analysing them in and through digital media. Many projects on ICTs appear to steer away from discussing the changing roles of production and practice in both learning and in research. Both dance and digital media offer possible ways to understanding how to research and to communicate about performance in ICTs, culture and communication. In this book, we do this via print technology and a set of established discursive conventions. However, as the appearance of images and weblinks suggests, this chapter refers to another, digitally mediated, communication platform, namely the Web. In our presentation of this chapter at the accompanying SKIKT conference,7 we further draw on oral performance as well as the Web to suggest ways in which communicating research might be conducted in addition to written expository discourse. For us this is not merely a rhetorical game: it is part of performing and practicing a wider communicative context for ICT-related research. In general terms, this is a constitutive research discourse in which digital media is both object and subject, synthetic and analytic, medium and message.

Finding our feet

A collaborative, interdisciplinary improvisation

In summer 2001, a cooperation between the SKIKT funded KTK project called Assemblages, based at InterMedia at the University of Oslo and the Departement of Ballet and Dance (Statens balletthøgskole) at the Oslo National College of the Arts was initiated.8 From the start, a process-oriented and improvisational approach characterised the collaboration. What emerged was an experimental multi-purpose project which aimed to link dance, digital media, education and research.9

Researching performance, performing research

One of the objectives was to integrate digital media and dance, creating a hybrid performance. This piece would be shown in front of audiences at our two institutions before the Christmas holiday. We also wanted to use ICTs in experimental teaching and learning as an inquiry into how ICTs could be used to generate institutional change, and to research how ICTs could be informed by dance and choreography. This chapter is therefore one of a series of publications generated from this project (see e.g. Morrison et al. 2001, the Ballectro project website 2002).

Choreography meets digital media design Cartography. Mapping. Dancers working with computers know that mapping is not the old explorers' dream of discovering a terra incognito. We do not confront a new territory, but dance through a transformation of exisitng material realities and relations. Once intitial squeamishness is overcome, dancers make the best cyborgs. Haven't we been shaping and distorting our bodies and abilities with a variety of techniques for centuries. Dancers know that the borders of the body are mutable, porous. When dancers engage with systems like motion capture they crave close contact with the abstracted digital data, not to annul it, but to share in contrasting spaces and physicalities. (Susan Kozel 1997).

This quotation points to the importance of the choroegrapher, Jane Hveding, with whom we worked on Ballectro. She works as a dancer, a freelance choreographer and dance teacher. She provided us with an experimental, 'free-form' approach to building a collage-like choreographic process. In addition, she invited us into this process and provided us with the security in which to improvise as multimedia makers, teachers and researchers. This offered us a context within which to investigate the reciprocal relationships between performance, movement, space and expression in digital media and dance. Importantly, this learning context was one in which we were not overshadowed by a dance professional who insisted that her views dominate. Nor did she see the project in terms of more established approaches to film/video and dance. This provided us with an important 'space' in which to start 'finding our feet' together.

Dancing on the academic table Jane Hveding had been selected by Statens balletthøgskole to assist six final year dance students in a course on choreography in the fall of 2001 leading up to the annual student performance. We used parts of

221

222

Researching ICTs in Context

the summer 2001 to meet with Jane and to discuss how such a collaborative project might be shaped and carried out. As none of us had yet met the students, we used this time with her to experiment with the facilities in InterMedia's new building.

Figure 2: Jane Hveding in the video conference room at InterMedia, summer2001 on the first day of our collaboration; (left) infinite regression in projected image; (right) the same event filmed from a different camera and projected onto one of the other screens. We began this by workshopping in the video conference room, perhaps not the first space to think of in terms of a dance and digital media project. By introducing a choreographer to this facility, we suddenly found ourselves in the middle of a 'stunt' performance, where the videoconferencing system was transformed into a surveillance camera arena for projecting the moving images of the choreographer who had begun to dance on the room's tables. Right at the start, we could see, and we could sense, that physical movement and improvisation would be central to the project. Not only would it be a way of trying out different choreographic configurations, it would also be a way of thinking about how the 'apparatus' of digital media might itself be moved, shifted off the desktop and used reflexively to suggest ways of looking at movement and the visual, as dance, as digital media and as digital-dance. The workshopping moved on to InterMedia's tv-studio buried in the basement of our new building. This was a studio still very much needing to be used and promoted and, therefore, later we were able to comandeer it for weeks at a time. Initially, on our part as multimedia designers

Researching performance, performing research

and researchers, we saw this not as a tv-studio, but as an experimental multimedia space. However, this was not how the studio had been designed, nor was it how the technician responsible for it viewed how the space ought be used or managed. Over time, her interests in the project grew and she made an invaluable contribution to its technical success. However, at first she was alarmed that we wanted to unscrew video projectors, remove lights and replace enormous, professional tv cameras with an improvisation process including hand-held digital video cameras. In this project we would repeatedly need to explain that we were not interested, as it were, in pre-defined settings, either spatially, technically, choreographically or educationally. We would need to explain that we were learning how to collaborate in an interdisciplinary performance-based project. We would again and again also need to explain that the nature of this developmental production and performance research inquiry would need to be understood in terms somewhat different to many of the reports on ICT and learning projects in which research designs are already largely given.

Through a glass darkly In one of the first shared sessions in the studio, which included the choreographer, technician, and multimedia makers, we experimented with equipment which was already available. Here we were interested in seeing the space as other than a zone for broadcasting. A computer projector, a mobile fabric screen and different lights were combined in ways that pre-figured the collage-like structure of much of the project. What was also important at this early stage, as Stone (1995) suggests with digital media, was that we improvised and played. In this studio we rarely sat down, other than on the floor. In this session we found an ordinary drinking glass which somebody had left in the room, and as a result, the blue light from the projector (not yet connected to a laptop) was beamed through the end of this glass. In this play, we took turns in moving between the projector and screen, and behind the screen, some of the effects of which can be seen in Figure 3 below. We saw for oursleves that we would need to think through the work with the students, and for the final performance, in new ways for dance and for digital media. We would need to rethink more established conventions of video and film on and for dance, just as for digital media we would need to move beyond the confines of the desktop.

223

224

Researching ICTs in Context

Figure 3: 'Through a glass darkly …'. Andrew Morrison and Jane Hveding during the first session in the tv-studio at InterMedia, summer 2001. The inverted camera was used together with video feedback in the later performances. These examples are mentioned because they illustrate a bottom-up way of working which was essential for the project. The playful mood in which this collaboration started was continued during the further work with the students and with making the performance piece itself.

Reconfiguring learning contexts Collaborative learning

To work on such a project of simultaneous teaching and learning, we needed to learn about dance and choreography. We also needed to learn how to incorporate the media in a performance context. In broad terms, we had to learn by doing and this doing was to be done via workshop dance and digital media sessions in which improvisation, play, experimentation and rehearsal were central. In this respect, the project drew on the notions and actions approach of the reflective practitioner (Schön 1983, 1987) and tried not only to apply them to dance education (McFee 1994), but to the composition of a multimedia-dance project: as learning, as design and as research. Where such an approach may be opposed to a technical rationalism, it sees knowledge as inherent practice and practice as a means of finding solutions to problems, as well as theorising them. Thus reflective practice is a route to generating knowledge, but one in which this is seen as socially constructed. In the case of the Ballectro project, this was an in-

Researching performance, performing research

determinate, processural reflection-in-action, and one that needed to look outwards from known disciplinary borders in the collaborative shaping of a new hybrid performance and space. As Wei and Kuzmanovic (2001) comment on the context of their collaborative dance and media work, 'It is the rich confusion of our physical world, together with the instability of the virtual one, that allows hybrid public spaces to emerge.'

Meeting the dance students At the start of the fall term in 2001, we met with the six students at the Statens balletthøgskole. The dance peice would need to be ready by late November. The students were already accomplished dancers after two years of full-time study at their institution. They were six very different people, if of roughly the same age. The group included three women and three men. Two of the men, Koshiwayi Sabuneti and Wilson Phiri, come from Zimbabwe. Two of the women are also from outside Norway, Beáta Kretovivová is from Iceland, Malin Rengstedt is from Sweden, while the third male dancer, Erlend Samnøen is from Norway as is the third women, Stephanie Sund. English and Norwegian were therefore the working languages of this diverse group. These students had expressly asked for inputs on video and dance and were interested to learn more about how digital media might be used in dance, though they had themselves not moved much beyond web browsing and SMS messaging. We introduced the students to the project members and to a new institutional and practice/performance setting at the University of Oslo. In addition, we developed a tight schedule for collaboration with them and their choreography teacher. We soon learned that, given their other dance commitments, there would be little time for training in digital media. All-the-same, the students were to play an active part in making the performances, and many of the ideas and movements in the workshops and the final performance also came from the students. The negotiation between the various actors in the project — six dance students, a choreographer, two technical personnel, a musician, and three media designers/researchers — was complex. We were involved in a 'developing project' in which delicate negotiations were repeatedly needed between persons, cultures and disciplines. The altering of traditional roles such as choreographer, editor, producer and designers and dancers led to interesting discussions among ourselves when we had to decide the credits

225

226

Researching ICTs in Context

in the programme for the Christmas performances. The dancers had made their own solos, the project leader was doing media design and the engineer worked on the lighting, and the choreographer was just as much a project leader. The researchers were doing choreography. As is the case in such collaborative projects, we often took on slightly different roles and had to learn to function within and beyond our own fields of expertise. In hindsight, though, one could argue that a stricter, more linear way of working might have been more effective in developing a finished 'multimedia' production. Had the logistics of the project been more thoroughly anticipated and planned, our medley might also have been easier to perform. ICTs in the learning context of Ballectro meant that both teaching, learning, making a performance and researching were different aspects of the same project. Interdisciplinary collaborative projects which involve experimental new media have a tendency to turn established roles of teachers and learners upside down. In such an experimental project, nobody is an expert, and everybody has to learn from one another to be able to the work collaboratively. In the Ballectro project, stepping out of one's own professional roles was required for creating a climate for collaboration and workshopping. From the media side of the project, it was necessary to learn about dance, choreography, and performance to be able to think about the media as an integrated part of the performance. The choreographer and dancers had to learn about digital media to be able to see possible relationships between the dance and the media.

Practice and performance spaces The actual bodily movement of real people, and not only the screen-based design of digital media, was an important dimension in the collaboration. On many occasions the choreographer answered our questions by dancing, by moving, by using gesture and by drawing. We gradually learned — by her example, and by watching the students also moving and learning — how central thinking with the body is to choreography, dance design and performance. This was most important in rethinking our own understanding and design of digital media in the learning processes of the project and in the performance context. As will be discussed later in the chapter, this also further influenced our approaches to communicating research online.

Researching performance, performing research

The movement in an actual physical space (the large dance stage at Statens balletthøgskole, or the Intermedia tv studio with its smaller 'stage' area) provided the research project with a means of bringing design and research of ICTs into several interesting spatial relationships and contexts. The two performance spaces differed in size and quality. As far as dance was concerned, the larger professional stage at Statens balletthøgskole was superior, but it needed to be booked and was in great demand. In contrast, the cold concrete floor of the tv-studio, now called the studio, was more of a private rehearsal space where it was possible to leave equipment which had already been set-up. At various points in the project, the multimedia-related equipment had to be hauled between this studio and the stage at Statens balletthøgskole, and re-purposed for that different context. So much for 'mobile media' in all this shifting of equipment in a project about movement! We have described and discussed some of the media related components of Ballectro in an online paper (Morrison et al. 2001). Here we refer to three examples only. At Statens balletthøgskole, we tried out interactive Flash animations with different movements and shapes.10 The animations were projected at the back wall on the stage, and the students were asked to select items from the computer and to interpret their movements and to improvise on the basis of them. Further, these selections were randomly generated by the software. Some of the students felt that the movements in the animations restricted them instead of giving them material for new movement. Given time constraints, these activities were not followed up in the way originally planned: we tended to move along with the dance at this point. However, some of these animations were used in the final performances but as scenography to a choreography inspired by the postcards exercise described at the start of the chapter. We also introduced a projected live camera to the teaching. The camera was placed upside down at a certain angle to the back wall. This created a visual feedback effect. The students needed to find out how their movements could work together with the video. They quickly found ways of moving — ones that were clearly also producing and developing a strong aesthetic — that they probably would not have found out without the feedback. Example of this can be seen in Figures 4 and 5 below in the form of Willson Phiri's boxing-match with his upside-down self, and Erlend Samnøen's generation of his own dance moves from seeing them multiplied in front of him.

227

228

Researching ICTs in Context

As with the animated shapes in Flash, this video feedback was developed into a part of the final performances. So too was the use of 'spatial video' in the live filming and projection of close-ups from group and duet sequences. Auslander (1999) has discussed 'liveness' in performance and media. In the actual final performances we wanted to maintain a feel of the importance of improvisation and liveness, of its performativity. In The Analysis of Performance Art, Howell (1999: 229) argues that: A key question for performance artists must be, how do we keep the improvisation diverse, how do we ensure that even our most spontanesous actions read ambiguously, so that our audience finds it difficult to decide whether it is watching something improvised or something rehearsed.

Repeatedly, we found that the relationships between movement and the media were strongly interdependent, and there was a knack to their cocomposition.

Figure 4 & 5: (left)Willson Phiri boxing with himself at the first upside-down camera session at the Ballethøgskole, fall 2001. (right) Erlend Samnøen at the first upside down camera session at the Ballethøgskole, fall 2001. Notice the feedback effect on the graphics from the camera display. We also invited students to film with DV cameras while they danced. We wanted to try out how they would move with the cameras, as dancers and not from the standpoint of more traditional video. We were able to see how this dance looked form the point of movement of the dancers; this would be inaction, inside footage. We used this event to suggest to the students that they should be thinking of how to 'see through' other aspects of the use of digital media in the project. We also suggested that this would be a likely need in understanding and practising the future interplay of dance and technology.

Researching performance, performing research

Figure 6: (left) Malin Rengsedt and Beàta Kretovivovà film a duet between Stephanie Sund and Koshiwayi Sabuneti; Beàta Kretovivovà films a solo imporvisation by Erlend Samnøen. This inverting of camera and dancing with images of live images point to how important it is not to take technology and tools as a given, but to see them as felxibel and mallable, despite their boundedness. Similarly, if interdiscilinary teaching and research into ICTs is to succeed, it is important to let go of a certain degree of certainty and to step forth with some security given by a collaborative partner.

Performance and context

One 'performance', two contexts As the project developed, we decided to use the studio at InterMedia and the stage and rehearsal studios at Statens balletthøgskole as a performance spaces. This decision was made for practical reasons: the dancers needed room in which to keep developing their project and we needed a stable testing ground in which equipment did not have to be removed from the practice space immediately after the session. This was possible at InterMedia, yet the studio venue there was not designed as a dance space: it was smaller than the stage at the dance school, and its hard floor was not kind on dancers' feet. However, the studio provided the project with its own space for rehearsing, and for accentuating the place of digital media and research in the project. Further, this brought dancers into a research setting, and also allowed colleagues from the department to see our work in progress, as dance, as digital media and as an exploratory research project. The decision to move between two unalike spaces was also made because we intended that the work also be performed in two different contexts, one connected with research and the other dance.

229

230

Researching ICTs in Context

Production with research However, this shifting context in which the project was being developed also added extra levels to an already complex project process. In the research context around InterMedia, the project was not easy to categorise and, with a very small budget, technical needs had to be redefined. The project could only proceed on the basis of many technical compromises. While this was at times frustrating for all the participants, the project was designed as an exploratory learning and design collaboration between new partners over a short period. In addition, we were working to develop a student performance piece in which digital media was not meant to over-shadow their dancing. In the context of Statens balletthøgskole, we felt that our piece was seen more as a danced choreography than as aa dance-media hybrid. The project did not fit easily into the existing structures of the two institutional contexts. Yet, as a whole, the project was useful in piloting the needs of such action-based research in a new research centre, where many facilities had yet to be used experimentally, and where working with digital media production and research into electronic arts had not been part of the main concerns of either research or production. The project also brought research on choreography and digital media into the dance school.

Performance for research Improvisation for researchers

One of the difficulties of this project was explaining the importance of improvisation and experimentation in the medley of dance, learning and digital media: as learning, as design and as research. We decided one way to do this was to bring the project out of the studio in the basement of InterMedia and into the central public space of the building. The six students improvised dance in this space for approximately fifteen minutes. The space includes the entrance and the canteen as well as a flight of three levels of stairs open to the other areas, and linked with glass elevators. This offered people many points from which to view the improvisation. It also allowed the dancers to move across and up and down public spaces and to draw our attention to our own typical movements across them. The improvisation took place without music and against the backdrop of the daily activities of this pubilic arena. The students ran up and down the stairs, following some pre-arranged moves. At times they followed a

Researching performance, performing research

movement one of them had begun. They fell to the floor. They wrapped themselves around the bannisters and descended the staircase in a large moving mass. One of the students raced up to the top of the stairs way above the onlookers and shouted out into the roof of the glass atrium. Others danced in and around the elevators, pressing the buttons, holding open the doors and breaking the rules of polite public use. People going about their business walked right into this improvisation and responded to it differently. The leader of an ICT and learning project was physically drawn into the dance. As the improvisation piece came to a close, the large group of onlookers began to return to offices, labs and meeting rooms. The student dancers also introduced improvisation at an interval in a seminar on Mobility held at InterMedia in November 2001.11 The students had been invited to perform an improvised piece to highlight aspects of human movement and artistic expression through dance. On this occasion, however, they performed in the open air, in a large courtyard hemmed in by glass walls and set into a well in the main entrance level of the research park. Participants in the research seminar on mobility were asked to leave the warmth of the seminar room and to move outside to see a dance piece on mobility and improvisation. As part of their improvisation, the students used the air streams from a large ventilation system cone, and blew leaves, paper and other objects into the air. The audience, looking down on the performers, was encouraged to throw chocolates, paper balls and empty cardboard boxes down into the space. The students then reacted to some of these immobile objects. They also took photographs during their performance. These were later used in an installation the students made in the canteen at Statens balletthøgskole as part of their contextualising of the project for their end of year performance.

Performance for researchers Having introduced Ballectro as dance to a research context through improvisation, the project was now introduced as dance and media to teachers, graduate students and researchers through a research seminar entitled 'Designing Design' which we also arranged. This seminar discussed how different fields of design may inform research and development in digital media, and how we may employ approaches from design to investigate, analyse and understand digital technologies and their uses. It also asked in what ways an inte-

231

232

Researching ICTs in Context

grated approach to design can help us to conceptualise and practice the use of ICTs in communication, learning and culture. Papers for this seminar were published in electronic form only. The media researchers in the Ballectro project used this occasion to present the project from an online research paper. The paper concentrated on collaboration and the role of digital media in the project (see Morrison 2001). For us, the overall seminar was a means of building an electronic resource of academic publications around design and digital media. We were able to present material on the media components of the project as well as its collaborative and participatory design. This 'design of design' was then extended at the close of the seminar with the first full public performance of the dance piece. The researchers, then, shifted their roles from the seminar and became performers of a different kind (mixing sound, video, projection, and computer files).

Figure 7: Video still from a trio, performance at the Designing Design seminar, InterMedia,28 February 2001; dancers: Malin Rengstedt (back) and Koshiwayi Sabuneti (front).

Researching performance, performing research

Lasting roughly 20 minutes, the performance was specifically for the seminar participants and other interested researchers. This gave these researchers the possibility of 'performing the text as an audience', and not reading about it in a report or article form. Video material of this performance is to be found on the Ballectro website under development. We invite readers of this chapter to interrupt their current role as readers of print academic discourse, and to see a mediated form of the performance by visiting the 'Performances' section of the Ballectro web. This interruption is a deliberate part of our wider argument about the possibilities to move between layers of discourse and discourse types in research which is about and in digital media and performance. Our presentation of this chapter at the SKIKT conference, for which this book has been prepared is a further attempt to cross boundaries between print and electronic discourse, between presentation and performance, and between production and interpretation.

Figure 8: The studio space at InterMedia after the Ballectro performance at the Designing Design seminar, 28 November 2001.

233

234

Researching ICTs in Context

Performance for dance

From research to dance context Following these improvisations and the performance for researchers at the University of Oslo, we now moved to the performance setting of Statens balletthøgskole. This was a shift in context, in terms of space, technology, performance and audience. This shift also impacted on our understanding the mechanics of the project as well as the interpretation of the performance piece. While Ballectro was given prominence by being the opening piece in the annual student concert, it was very clear to us as media developers and researchers that we had entered an entirely different learning and performance settting. This proved to be important in two main respects. The first concerned technical adjustment. The second refers to the multi-level choreography and multimedia character of the piece being placed in an established dance performance schedule. Moving to the larger and wider stage, and the fact that our performance would be replaced by another scenography, meant that we had to reconfigure the entire technical arrangement of the work. Although we knew the dimensions of the stage from having rehearsed there, we had to alter the screens we had made to fit this space. However, we did not change their size as we would have liked to have done. We also quickly realised that due to the scale of the performance space having expanded, the dance itself would open out. Similarly, our use of digital media would need to accommodate a broadening of movement. Most importantly, this meant that a second, more powerful computer projector had to be added to the technical repertoire. This was so that the computer projected material would be bright enough to be seen by a nightly audience (of about 150 persons). This audience was also likely to be more critical of the dance and choreography, and the ways in which multimedia enhanced or detracted from them. As the piece was to be followed by an entirely different dance, the set for Ballectro, including expensive computer equipment also had to be taken down in minutes by the non-dancers in the project. This was quite a different kind of performance for us as researchers to conduct over six nights. While in its processes of composition the piece had been frequently restructured, now we repeatedly re-performed the same structured piece.

Researching performance, performing research

Figure 9: Stills from a Ballectro performance at Statens balletthøgskole. The dancers (above) are multipled on the back screen via video projection of live camera feed. The dancer (below), Beàta Kretovivovà, observes the projection of a previously recorded digital video of herself dancing. In Icelandic and Slovenian, she questions her own moving self. Simultaneously, music (compressed and accelerated sound built from her own voice) plays. While so far we had called the piece and the project Ballectro, suggesting a merger of dance and electronic media, the students and the choreographer now decided the piece needed a different, and less bal-

235

236

Researching ICTs in Context

let-like name. After much debate we settled on the title 'flæsj frash'. This pointed to two aspects of the performance. The first was that we had used the software Flash to develop part of the dance, but of course the possible 'Flashdance' was the title of an entirely different Hollywood film from the 1980s. Second, the title 'flæsj frash' referred first to the Norwegian spelling of Flash and, in a playful other senses, to one pronounciation of the word by a Zimbabwean speaker of Shona suggesting 'coolness'. This allowed us to refer to the Norwegian and Zimbabwean participants in the project, though the nuances in this title may be partly missed. Another change we met was that our formerly contained performance piece now joined a medley of others in an established dance performance schedule geared towards a dance community. In this context, we were also finally able to see our work in relation to other choreographies, some developed by the staff and two based on the work of the well-known Norwegian chorographer, Jo Strømgren. An additional, and difficult, change for us as researchers who had driven the digital media part of the performance was that in this performance context some of the technical control now needed to pass into the hands of a lighting and sound specialists contracted by Statens balletthøgskole. This meant, especially as regards sound, that we were no longer able to integrate all the elements of what was a multi-level choreography and multimedia performance ourselves. Used to mixing thes elements in tune with the dancers and other media, we now saw how separate many dance performances are from their creators. Perhaps this is no surprise to choreographers and lighting and sound specialists.

Performing research

Reporting research in online environments There is a small, if growing body of research about conducting online research (e.g. Jones 1999, Mann & Stewart 2000, Hakken 1999; see also this volume). This print based material is chiefly concerned with how to carry out a range of research methods in primarily Internet based domains. Applied research is presented but little mention is made of visual media (see e.g. Pink 2001) or of building a multi-mediational research rhetoric. Such an argument is to be found in several online journals concerned with scholarly publication and digital discourse (e.g. Gailin & Latchaw 1998, Ingra-

Researching performance, performing research

ham 2002). However, as Morrison (2001) has argued, few electronic journals in the humanities and social sciences make apt use of images in their web-based mediation of research. In addition, fewer still employ video material as part of an electronic research rhetoric (e.g. Owens 2000). Digital video and audio were used to document aspects of Ballectro. Still and moving images were an important part of reflecting on the learning and work process as well as the performances.12 We used the projection of recorded and projected video to help us decide how to shape the larger choreography; reviewing footage also helped in selecting and refining material for the piece. Drawings, sketches, and notes were used to keep track of a complex process. For some of the period a sociologist observed our sessions, and although given the density of the project his role as a video documenter was not fulfilled, he nevertheless introduced important questions about participant observation in a performance-driven project. Dance itself was also a mode of memory-making: the repeat movements, the rehearsals and the need to 'learn' the piece as non-dancers forced the multimedia developers to put down their notebooks and cameras. In addition to the video material filmed by the researchers, at times the students also filmed their own dance movements as well as the contexts in which their learning was taking place at Statens balletthøgskole. Video was also used in several workshop activities to highlight issues of presentation, and improvisation, though video-playback of dance was rarely used. Elements of the dance were repeated with changes and different expressions. This reflexive development design also filtered into the ways in which we also worked as participants and as participant observers ourselves. However, face-to-face discussions and actual 'rehearsals' in the workshop sessions were the most frequent means of making sense of the ongoing process. The aim of using digital video was not to produce a fully documented multimedia ethnography online, though such tools now enable us to follow research and learning trajectories via the web. However, building a multimedia-rich website (now underway) was one way of documenting aspects of the project: as a means of record for the dance school and as an aid to future learning and research projects (Ballectro 2002). We were also trying to perform research rhetorically, by way of experimenting with how to present it in an online setting. Thus we aimed to generate an example of how one approach to projectbased research may be communicated in a digital domain. In particular, as the Ballectro website demonstrates, we needed to find a fit between the

237

238

Researching ICTs in Context

content, the process and the medium so as to reduce some of the distance between a partly digitally mediated performance and its interpretation.

Extending performance based research This chapter, therefore also, lies within the evolving Ballecto project website. It too is part of a wider project at InterMedia to build capacities for designing research materials for online communication. Not only would we need to establish how and where and why to use still images, but these questions would also apply to video, and the quality, size and streaming of files. Thus, we needed to select material from the process and the performances. We linked them with research papers, such as this one, as well as to other research on dance and digital media available online. In short, we were trying to develop a prototype to demonstrate how the web is a medium within which a convergence between the research processes and products may be constructed, and thus also interpreted. Jones and Stephenson (1999: 8) argue that: Interpretation is, we would argue, a kind of performance of the object, while the performance of the body as an artistic practice is a mode of textual inscription. The body (as the corporeal enactment of the subject) is known and experienced only through its representational performances — whether presented 'live,' in photographs, videos, films, on the computer screen, or through the interpretive text itself. Interpretation, like the production of works of art, is a mode of communication. Meaning is a process of engagement and never dwells in any one place.

With respect to Ballectro, such 'interpretation' may be seen at two levels. First, the project may be read as an instance of what we call 'performance based research'.13 Second, we suggest that it may be useful to broaden the electronic reporting of media-rich ICT-related research to include performance.

Future steps

Electronic rhetoric and hermeneutics In researching about dance and digital media, we were somewhat surprised at what we did not find online. While many dance sites exist, including those which discuss dance and technology (e.g. the Dance &

Researching performance, performing research

Technology Zone), few of them have high quality images. Rarely does one come across a dance site filled with images and text which contextualises these images.14 We imagined that we might also find video material on dance online, but this was even harder to find.15 There would still appear to be relatively few examples of dance online in which digital media is prominent but also where a multimediated dance may be said to be represented (see e.g. deLahunta 1998a, Birringer 2002) and analysed (Sha & Kuzmanovic 2000). Online examples of the documentation of dance in context, such as that from the Department of Dance at Ohio State University (2002) also exist, particularly with reference to dance and digital media workshops (Birringer 2002). Robbie Shaw (2002) has a five and a half minute web dance video called 'Time train'. Re-mediating the genre of a silent film, the piece has chorographic input from its performers. Further, this webscreened video was edited digitally. In another work, called 'Panic skid', Jennifer Marshall and Robbie Shaw (1999) present two dance videos which run simultaneously and which are surrounded by text. The documentation of electronic arts projects is an important part of building its history and its rhetoric (e.g. Dépocas, 2001). However, unlike many of the performance pieces of the 1960s and even 1970s, digital video, stills and audio now offer handier and cheaper means for documenting performance-based projects, and especially the corporeality of dance, and their mediation online. There are now several projects on the documentation of choreography and related dance performances. The Norwegain dance and multimedia performer Amanda Steggell and her collaborators provide a close-to-home example of how a performance work may be viewed online; in 'Maggie's Love Bites'16 the online mediation is part of the message (see also deLahunta 1996). This is well illustrated in a cd-rom based documentation of choreography for work entitled 'A Desparate Heart'. Similar is an instructional cd-rom 'Prey: an innovation in dance documentation' on the choreography of Bebe Miller (Mockebee 2002). This cd-rom contains notes, each of which … displays Bebe teaching or coaching the dancers alongside the Labanotation phrase, a complete edited video of the group work, historical, cultural contexts, interview with Miller, review by Candace Feck, Miller's World, and brief instances of the process and moments of the original long phrase that Miller and the students created.

239

240

Researching ICTs in Context

Such an approach is also taken over to dvd in a title called 'Going to the wall – a document of the process', again referring to a work by Bebe Miller. Such experimental research and performance settings provide us with a 'laboratory' space within which to conceptualise the changing character of dance and digital media. In particular, they also provide us with research driven contexts based on practice linked with theory. These contexts provide actual, populated interdisciplinary spaces for the embodiment of the synthetic-analytic model Gunnar Liestøl (2001, 2002 forthcoming) proposes is so important in digital media production and analysis. 'Peformance' may thus be seen to be extended beyond the role of the dancer and the static, seated audience.

Responsive digital media and performance 'Performance' may thus be seen to be extended beyond the established and assigned roles of the dancer and the static, seated audience to a more plastic relationship between tools and tellers, performers and the audience as performative players who may enter a performance space and piece and traverse the divide between the stage and the spectator. This can be seen in the collaborative works of Sponge and FoAM (sic). These are interdisciplinary and collaborative research and design teams working with digital media. 'Tgarden' is one of their interesting performance based project wesbites (Sha & Kuzmanovic 2000, Sha et al. 2000).17 This site, and their others, presented us with prototypical examples of performance, including dance, digital media and web mediation are co-present. Importantly, the 'Tgarden' project site stresses the importance of responsitivity in digitally mediated performance. A collaborative and interdisciplinary research, technical and artistic venture, 'Tgarden' investigates: … how people individually and collectively make sense of responsive, hybrid media environments by articulating their knowledge in non-verbal ways. More specifically, the project investigates how a person can create meaningful gestures in a dynamic environment and develop expertise in them.18

This can be seen in the videos which accompany this explanatory verbal text. In a section of their website entitled 'Transforming the tool', Sponge state that 'We're interested in making it possible for someone who is trying to "write", in the broadest sense, to refashion the tools of writing him or herself. A reflexivity of action.'

Researching performance, performing research

Sponge and FoAM presented us with digitally mediated research discourse in which digital media was employed to communicate the research content and context. In short, it demonstrates a web-related contextualisation of interests and publication types similar to ours. In their own words, the web site is a hybrid and augmented space, but not just about performance but also as the performance of their research. This is apparent in their publication of a conference paper bearing video material from the performance (by the audience) as well as slides from the presentation of the paper. Two of the project members, Sha and Kuzmanovic (2000), argue that 'By shifting attention from representation to performance, we shift the focus of design from technologies of static representation (e.g. snap-shot database schemas with data from forms), to technologies of creation and performance.' As deLahunta argues, (and as we have experienced as learners, designers and researchers), arts education, including dance and media, needs to be broadened to include the problems and approaches we have tried to contextualise here as part of an ongoing shift in our literacies. He demonstrates this himself in an online piece as a 'temporary typology practice' referring to dance and digital media. He links a presentation text with its presentation notes, video footage and web links (see deLahunta 1998a).

Dymanic perfomative discourse As Susan Kozel (1997) says, 'How do we map the body of dance as it expands its representational systems?' How too are we to do this for digital media? That is for dance and media as artistic and as academic texts? Now that dance cds and dvds and research projects and publications have begun to demonstrate links between choreography and its performance, we hope that it will be possible to stretch this further and take some new steps into a dynamic, electronic and responsive performance. This may be the case in a proposed collaboration with three choreography students in the fall of 2002 in which they will each develop a choreography for other students to dance. At present we are still to learn about their interests, but their chreography teacher has asked us to collaborate with them in working on renaissance material. At present we are calling this the Piazza project. Piazza might be one site in which we can further investigate relationships between dance and digital media design, and also perhaps learn how to work and to develop more responsive multimedia dance spaces and performances.

241

242

Researching ICTs in Context

From the point of view of research and pedagogy on digital media, through the Ballectro project we learned to perform, to research performance and to perform research. A future interdisciplinary and collaborative multimedia-dance project such as Piazza might result in 'a site' in which research processes, accounts, performances and interpretation are connected with an electronically mediated rhetoric. In such future steps, we would also hope to embody the performing of research so that 'moving media', responsive performers and audiences might also be 'players' in its making and interpretation, stepping between digitaldance and digitally mediated dymanic research texts. That, however, is a matter of a different time, space and context.

References (All references and notes related to the World Wide Web are cited on 19 March 2002). Aarseth, E. 2002. 'Computer game studies, year one.' Editorial. At: http://www.gamestudies.org Auslander, P. 1997. 'Ontology vs. history. Making distinctions between the live and mediatised'. The 1997 Performance Studies Conference. At: http://webcast.gatech.edu/papers/arch/Auslander.html Auslander, P, 1999. Liveness. London: Routledge. Avison, D. 1997b. 'Action research in information systems'. In McKenzie, G., Powell, J. & Usher, R. (eds) Understanding Social Research. The Falmer Press London: The Falmer Press. 196-209. Birringer, J. 1996. 'Lively bodies-lively machines'. Workshop report from the SPLIT SCREEN Festival. At: http://www.art.net/~dtz/birring.html Counsell, C & Wolf, L. 2001. Performance Analysis: an introductionary coursebook. London: Routledge. DeLahunta, S. 1996. 'New media and information technologies and dance education'. Paper given at international symposium Future Moves, Rotterdam, Sept. At: http://www.art.net/~dtz/scott1.html deLahunta, S. 1998a. 'Sampling ... convergences between dance and technology'. Paper presented at the Art Crash Symposium, Aarhus, Denmark, 2-4 April. At: http://www.art.net/~dtz/scott2.html

Researching performance, performing research

deLahunta, S. 1998b. 'Speculative paper: theatre/ dance and new media and information technologies'. Paper and presentation, Working Groups on Dance and Drama, Research Group on Reorganisation of Professional Arts Education, Amsterdam, April. At: http://www.art.net/~dtz/scott3.html Denning, P. & Dargan, P. 1996. 'Action-centred design'. In Winograd, T. (ed) Bringing Design to Software. New York: ACM Press/Addison-Wesley. 105-120 Department of Dance, Ohio State University. 2002. 'Documentation'. At: http://www.dance.ohio-state.edu/%7Eshaw/documentation.htm Dépocas, A. 2001. 'Digital preservation: recording the strategy. The documentary strategy'. In Takeover: who's doing the art of tomorrow. Ars Electronica 2001. Vienna: Springer.334-339. Fusco, C. & Bustamante, N. 1999. 'STUFF: a performance'. In Jones, A. & Stephenson, A. (eds) Performing the Text/Performing the Body. London: Routledge. 237-254. Galin, J. & Latchaw, J. 1998. 'Heterotopic spaces online: a new paradigm for academic scholarship and publication'. Kairo, 3.1. At: http://english.ttu.edu/kairos/3.1/coverweb/galin/index.htm Howell, A. 1999. The Analysis of Performance Art. Amsterdam: Harwood Academic Publishers. Ingraham, D. 2002. 'Scholarly rhetoric in digital media (or: now that we have the technology, what do we do with it?)'. Pre-print under review. JIME. At: http://www-jime.open.ac.uk/00/ingraham/ingraham-t.html Jarvis, P. 1999. The Practitioner-Researcher: Developing Theory from Practice. Jossey-Bass: San Francisco. Jones, A. & Stephenson, A. (1999). 'Introduction'. In Jones, A. & Stephenson, A. (eds) Performing the Text/Performing the Body. London: Routledge. 1-10. Jones, S. 1999. (ed.) Doing Internet Research: critical issues and methods for examining the Net. Thousand Oaks: Sage. Kozel, S. 1997. 'Material mapping: review Digital Dancing 1997'. In Dance & Technology Zone. At: http://www.art.net/~dtz//kozel2.html Kress, G. 1998. 'Visual and verbal modes of representation in electronically mediated communication: the potentials of new forms of text'. In Snyder, I. (ed.) Page to Screen: Taking literacy into the electronic era. New York: Routledge. 53-79.

243

244

Researching ICTs in Context

Kress, G. 1999. '"English" at the crossroads: rethinking curricula of communication in the context of the turn to the visual'. In Hawisher, G. & Selfe, C. (eds) Passions, Pedagogies and 21st Century Technologies. Utah State Logan: University Press. 66-88. KTK seminar. 2001. Mobilitet. InterMedia, University of Oslo, 21 November. At: http://www.intermedia.uio.no/seminarer/mobilitet/top.html KTK seminar. 2001. Designing Design. InterMedia, University of Oslo, 28 November. At: http://www.intermedia.uio.no:16080/seminarer/designingdesign/Index.html Laurel, B. 1993. Computers as Theatre. Reading MA: Addison Wesley. Liestøl, G. 2001. 'Research into the development of digital media as an interdisciplinary field based in the scinece and the humanities'. In Liestøl, G. & Morrison, A. (eds) Tverrfaglighet og Digitale Medier. Oslo: InterMedia/UniPub. 41-55. Liestøl, G. 2002, forthcoming. 'From synthesis to analysis (and vice versa): topics of conceptualisation and construction in digital media'. In Liestøl, G., Morrison A. & Rasmussen, T. (eds) Digital Media Revisited: theoretical and conceptual innovation in digital domains. Cambridge: The MIT Press. Mann, C. & Stewart, F. 2000. Internet Communication and Qualitative Research: a handbook for researching online. London: Sage. Marshall, J. & Shaw, R. 1999. 'Panic skid'. Online dance pieces. At: http://www.dance.ohio-state.edu/%7Eshaw/panic.html McFee, Graham. (1994). The Concept of Dance Education. London: Routledge. Mockabee, V. 2002. Faculty webpage. Department of Dance, Ohio State University. At: http://www.dance.ohio-state.edu/%7Emockabee/ Morrison, A., Skjulstad, S. & Aaberge, A. 2001. 'Designing performance: "where to put your feet" in a digital duet'. Paper presented at 'Designing Design' seminar, InterMedia, University of Oslo, 28 November. At: http://www.intermedia.uio.no/seminarer/designingdesign/Ballectropaper/D esigning_performance.htm Nielsen J. & Tahir, M. 2001. Homepage Usability: 50 websites deconstructed. Indianapolis: New Riders. Owens, R. 2002. 'The black press'. The Journal of Multimedia History, Vol. 3. At: http://www.albany.edu/jmmh/vol3/blackpress/blackpress.html Pink, S. 2001. Doing Visual Ethnography. London: Sage. Schiphorst, T. 1996/1997. 'Bodymaps: artifacts of touch'. Cited in Dance & Technology Zone. At: http://www.art.net/~dtz/schipo1.html

Researching performance, performing research

Schiphorst,T. 1997. 'Body Noise: subtexts of computers and dance'. Computer Graphics, February. Cited in Dance & Technology Zone. At: http://www.art.net/~dtz/schipo3.html Schön, D. 1983. The Reflective Practitioner. London: Temple Smith. Schön, D, 1987. Educating the Reflective Practitioner. San Francisco: Jossey Bass. Sha, X.W. & Kuzmanovic, M. 2000. 'From representation to performance: responsive public space'. DIAC 2000. At: http://www.f0.am/publications/2000_diac/index.html Sha, X.W, Salter, C., Farabo, L. & M. Kuzmanovic, M. et al. 2000. 'TGarden'. SIGGraph Art Gallery Publication, July 2000, New Orleans. At: http://www.f0.am/publications/2000_siggraph/index.html Shaw, R. 2000. 'Time train'. Online dance video. At: http://www.dance.ohio-state.edu/%7Eshaw/timetrain.htm Snyder, I. 1998. 'Page to screen'. In Snyder, I. (ed) Page to Screen: taking literacy into the electronic era. New York: Routledge. xx-xxxvi. Sponge/FoAM. 2001. 'Tgarden'. In Takeover: who's doing the art of tomorrow. Ars Electronica 2001. Vienna: Springer. 76-77. Stocker, G. & Schöpf, C. 1999. 'Preface'. In Drukrey, T. with Ars Electronica (eds). Ars Electronica Facing the Future. Cambridge MA: The MIT Press. Stone, A.R. 1995. The War of Desire and Technology at the Close of the Mechanical Age. Cambridge MA: The MIT Press. Trondstad, R. 2002, forthcoming. 'Performing the MUD adventure'. In Liestøl, G., Morrison A. & Rasmussen, T. (eds) Digital Media Revisited: theoretical and conceptual innovation in digital domains. Cambridge MA: The MIT Press. Tyner, K. 1998. Literacy in a Digital World: teaching and learning in the age of information. Mahwah NJ: Lawrence Erlbaum Associates. Wechsler, R. (no date). 'Computers and art: a dancer's perspective'. Cited in Dance & Technology Zone. At: http://ourworld.compuserve.com/homepages/palindrome/term.htm

Notes 1

Ballectro is a project funded by the KTK initiative, housed at InterMedia at the University of Oslo. KTK stands for 'Communication: Technology & Culture', a programme funded by SKIKT. According to original documentation, KTK 'focuses on the interfaces between the traditional subject areas. The central effort is therefore directed towards the development, application and

245

246

Researching ICTs in Context

analysis of communication technology in the light of historical/philosophical, linguistic/symbolic, societal, political, legal, ethical, pedagogical and technological factors. This applies to both the conditions for and the consequences of such technology. The aim of this priority area is to develop new competence through interdisciplinary research with a view to meeting the needs of society for new teaching provision, mediation, publicising and further research.' Statens balletthøgskole also contributed to the funding of the project. Information on Statens balletthøgskole may be found at: http://www.khio.no/ballett/index.html. The Ballectro project website is at: www.intermedia.projects/ballectro.htm. Our thanks to Bård Ketil Engen for comments on the chapter. 2 Statens balletthøgskole (known in English as the National College of Ballet and Dance) is part of the Kunsthøyskole in Oslo. In this chapter will use the Norwegian name. 3 Workshops are being used as part of the building of the Ballectro website; students from the project will be partly involved in this process. 4 Marita Liulia, Kimmo Pohjonen and Aki Suzuki's Performance Manipulator (at Kiasma, ARS01, Helsiniki, Finland) is a combination of concert, exhibition, media, and dance performance. It involves a similar mixing of art forms and ways of performing we found interesting in Ballectro. The Manipulator artists also combine traditional artforms (dance, visuals, music) with new technology. Available at: http://www.kiasma.fi/ars/manipulator/index.php 5 The 'Context' section of the PARIP website summarises this: The pursuit of practice as research / practice-based research (PAR / PBR) has become increasingly important during the past ten years to the research cultures of the performing arts (drama, theatre, dance, music) and related disciplines involving performance media (film, video, television, radio) as the contribution of the arts and cultural industries to national health and prosperity has climbed up the political agenda. A growing number of performing arts / media departments in higher education are now offering higher degrees which place practice at the heart of their research programmes. This represents a major theoretical and methodological shift in the performance disciplines — traditional approaches to the study of these arts are complemented and extended by research pursued through the practice of them. See: http://www.bris.ac.uk/parip/#context 6 See: http://www.bris.ac.uk/parip/#context 7 The SKIKT conference website is at: http://www.intermedia.uio.no/konferanser/skikt-02/skikt-research-conferance.html 8 For example, Tronstad (2002, forthcoming) suggests that the concept performance is only partially useful in conceptualising online MUD adventures.

Researching performance, performing research 9

Details of choreography and dance aspects will be addressed elsewhere, as will a focus on learning. 10 For more detail on the media aspects of the project, see Morrison et al. (2001). 11 This seminar and the one mentioned hereafter on design were both part of the KTK Assemblages project, funded through the SKIKT programme. 12 We documented parts of the process while the project unfolded. DV cameras were used to record aspects of the different sessions. Mostly the documentation was carried out as general observation without direct interviews of the students. The documentation is valuable for several reasons. It demonstrates a creative process but also miscommunication. We documented less and less towards the end of the project as we started to run low on people to fill the different roles, and do specific tasks. Another reason for the decline in documentation was the gradual change from a creative process to rehearsals. Towards the live performances more effort was put into making the performances actually work the way we wanted than on the recording of the process. Given our time boundaries, recording and interviewing the participants tended to break concentracion and to use up rehearsal time. 13 Here we do not mean the kind which is inherent in research assessment exercises such as conducted in British universities. 14 In contrast, for example, performance artists and groups, such as the Norwegian based Motherboard, have used the web to link real-time performances to websites with scripts and a range of different performance types to build dynamic environments. (see http://www.notam02.no/motherboard/). Troika ranch, for example, has a salon on projects (see: http://www.troikaranch.org/websalon2.html) and also provides a different image each day (see: http://www.troikaranch.org/yearbody.html). 15 Perhaps, the tradition of dance for film and the video of dance for television offer dancers higher quality images and more satisfying representations of choreographies and their performances. 16 From online notes to this piece: M@ggie's Love Bytes has developed through the group's engagement in net life, and tries to cater for a great amount, and diversity of traffic/inter-activity during the performance. The ongoing process of running the show is embedded in the expression of the piece (words such as open, close, fetch, connect, disconnect, and icons such as the running dog in FTP, and the open and closed eye in CU-SeeMe, appear on the wall, adding an extra dimension to whatever may be happening just then). All these features form an integral part of net life as we know it. And this is the point: M@ggie's Love Bytes is in fact a regular dance theatre performance of the 90's which dares to reflect upon, and perform through our digitally-connected lives as they happen - now! How-

247

248

Researching ICTs in Context

ever, what makes this project unique is the (re-claimed) power of the moving, gesturing human body in cyberspace, embodied in the dancers. At: http://www.notam02.no/~amandajs/where.html 17 Synne Skjulstad and Albertine Aaberge were able to become players in this responsive media and dance work at Ars Electronica in Linz, September 2001. 18 At: http://www.sponge.org/events_m3_tgarden.htm. See also: http://titanium.lcc.gatech.edu/topologicalmedia/tgarden/index.html

BLANK

11 Blogging thoughts: personal publication as an online research tool Torill Mortensen & Jill Walker

Introduction Once upon a time, weblogs were automatically collated overviews of data about visitors to a web server. That's changed. Nowadays the texts called weblogs are definitely not written by a computer. Weblogs today are subjective annotations to the web rather than statistics about it. Weblogs, or blogs as they are affectionately termed, are frequently updated websites, usually personal, with commentary and links. Link lists are as old as home pages, but a blog is far from a static link list or home page. A blog consists of many relatively short posts, usually time-stamped, and organised in reverse chronology so that a reader will always see the most recent post first. The first weblogs were seen as filters to the Internet; interesting links to sites the reader might not have seen, often with commentary from the blogger. Though weblogs have many different themes, looks and writing styles, formally the genre is clear. Brief, dated posts collected on one web page are the main formal criteria. Evan Williams, one of the creators of the popular blogging tool Blogger, is succinct in his definition: To me, the blog concept is about three things: Frequency, Brevity, and Personality. (..) This clarification has evolved over time, but I realised early on that what was significant about blogs was the format — not the content.1

This paper is about the use of weblogs in research. We are both researchers of online games, texts and culture, and most of our material is 249

250

Researching ICTs in Context

gathered online. A lot of our research is done online. Unsurprisingly, we came across weblogs when surfing the net. Discovering how simple Blogger makes blogging, we started our own weblogs. Jill Walker started jill/txt in October 2000, and Torill Mortensen started Thinking with my fingers not long after. The weblogs were originally used as a way to keep our focus while online, serving as constant little reminders of the real topics we were supposed to write about. They soon developed beyond being digital ethnographers' journals and into a hybrid between journal, academic publishing, storage space for links and site for academic discourse. Today our weblogs are among the most popular channels for frequent exchange of information between the members of what Mark Bernstein, Chief scientist at Eastgate Systems and writer of a weblog bearing his own name,2 dubbed 'The Scandinavian-flavoured cluster' of weblogs concerned with online communication and games.3 The generous spirit of blogging permits the writer to leave behind what Anders Fagerjord so fittingly names a Surftrail4 for others to follow through the World Wide Web, directing colleagues and others who might come by to areas of interest. And it's a trail annotated with everything from short comments, as is typical of Lisbeth Klastrup in Klastrup's Cataclysms5, to longer descriptions or even reviews, as is the style of Anja Rau's less frequent posts in her Flickwerk.6 Where home page remains a fixed noun, the word blog has rapidly become a verb as well. Bloggers have been likened to journalists, or perhaps better, editors; they might as well be compared to researchers. To blog is an activity similar in many ways to the work of the researcher. A weblogger filters a mass of information, choosing the items that interest her or that are relevant to her chosen topic, commenting upon them, demonstrating connections between them and analysing them. When we started our weblogs, we saw them mainly as tools for focusing, for exchanging information and being part of a discussion which potentially extends beyond the academic community. While Jill hoped to interest more people in the aesthetics of online stories and games, Torill deliberately used her weblog as an introduction to explain the research to players of games – potential informants – and let them follow the development of the thesis itself. This eliminated some of the mystery and tension related to research, and has on several occasions made it easier to cooperate with online role-players: the weblog establishes an accepted online presence which proves that the researcher is real to the digital space and not just a visitor with no knowledge. An personal online presence legiti-

Blogging thoughts

mates the online researcher much more efficiently than academic affiliation, flesh-world addresses or phone-numbers. To skilled online roleplayers, it's easier to fake a flesh-world personae than to maintain a consistent, long-term online presence. It also happens to give street credibility. Most importantly, our weblogs became tools with which to think about our research, its values, connections and links to other aspects of the world. They altered the way in we approached online communication, and have influenced the writing of both dissertations. This is the motivation for this article: a need to look at what weblogs do to our academic thinking.

Notes on our method Weblogs have become increasingly popular over the last two years, spreading rhizomatically to all spheres of the net: private, academic, cultural, professional, commercial and pornographic. An old weblog dates from the year 2000; an ancient one was started in the last millennium.7 Having participated in this movement, we have chosen to make full use of our own experience as bloggers and participants in the blogging community in researching this article, rather than conducting rigorous statistical surveys. This is a problematic position for a researcher under most paradigms: it defies the ideal of the not involved observer, and as such also contradicts the argument that research should be objective. It is however not an alien thought to the academic community, as Reason and Bradbury8 express in their description of the goals of Action Research: It seeks to bring together action and reflection, theory and practice, in participation with others, in the pursuit of practical solutions to issues of pressing concern to people, and more generally the flourish of individual persons and their communities.9

To the extent that it's possible to speak of online communities, bloggers create or maintain them. The post of Lisbeth Klastrup on the one year anniversary of her blog February 15th 2002 illustrates this: Also, quite a few bloggers have joined the community since I started and there is a nice sense of fellow exchanges and cross-postal reading between people like Jill (the mother-blogger), Torill, Hilde, Gonzalo, Anja, Adrian, Frank, Laurel, Elin, Carsten, and Lars. Anders is new on the blog, but will definitely be read. Jesper still refuses to blog, but at least now he is looking over our shoulders (*waves*). As is Ragnhild

251

252

Researching ICTs in Context

(*another wave*). And some of the Finns occasionally. Jan is a trusty reader too (*wave towards Jylland*). As is the omnipresent Mark Bernstein, who must be the most avid reader in the world ... (*respect*).

Since we are both part of these communities, and participate in, develop and discover them as we go, there can be no distance between researchers and the object of study, there can be no detached observation or cool clinical dissection of the topic. There would be very few weblogs of online research for us to study if we were to abstain from studying those of ourselves and this cluster. There is a considerable amount of popular writing on weblogs,10 but there is to date no published research on the topic, neither looking upon blogs as an aspect of digital culture, as a media phenomenon nor as a method or a publication tool for researchers. Likewise, not many academics write weblogs yet. When discussed in the media, weblogs are generally treated as belonging to popular culture or perhaps as being a form of folk journalism. There are many professional weblogs, though. As noted by early followers of the blogging movement, most of these are written by people who work online (web designers, software developers, and usability experts for instance). Researchers like ourselves who work with digital culture are then predictably early adopters of the technology. As professionals in more and more fields are starting to use weblogs we may see academics in other areas than digital culture enter the terrain. Several projects explore the use of weblogs in education, such as (obviously) Weblogs in Education, Edublog and Weblogg-Ed. These sites focus on the practical uses of online logs in teaching: the pedagogical effect of letting students publish and the potential gains in cooperation. At the time of writing (March 2002) they are mainly collections of links and examples of experiments conducted with weblogs, and are obviously works in progress.11 Many professionals keep weblogs, and they often use their blogs to reflect upon their work, to follow developments in the field and to publish ideas. Information architects12 and graphic designers are among the most prominent webloggers, and they often use their weblogs in a highly sophisticated manner. Much of the material processed and discussed in these weblogs is clearly research. University academics have been slower to adapt to the form.

Blogging thoughts

Statistics as to number of blogs and links to and between them are easy to find. One system for indexing links on weblogs is Blogdex, a project run by the MIT Media Lab in Boston. Blogdex gathers and analyses connections between weblogs based on links between them. From this information, the system generates lists of the most popular sites (popularity in the blogging community is here determined by links rather than by readers) and plots social networks between them. However, Blogdex isn't an absolute measure, and not all weblogs are indexed. Given the lack of previous research on weblogs, and of other researchers using blogs, we have chosen to base this paper on our personal experience, as well as placing our observations and experiences of the use of weblogs in general in a theoretical framework, drawing on media theory, rhetoric, reader-response theory and hypertext-theory, a universe of theoretical approaches which opens the article up in too many directions for us to cover within the scope of this work. We choose to focus on certain topics: online rhetoric, the style of blogs, academic weblogs: positioning research progression between public and private, and writing and thinking: how writing styles affect thinking styles. In many ways the questions we want to ask here relate to Michael Heim's approach to the ways in which word processing affects our thought processes. Our question assumes that weblogs are more than simple tools and that the way we write in a blog reveals something about how we think that would not be explicit in another medium. If we think of Blogger or Microsoft Office as a tool like a pen, 'to ask whether our thought process is affected by word processing is like asking whether using lead pencils will make my letters heavy and morose or whether blue ink will make my thoughts blue.'13 We are however already familiar with the hierarchy of pens and pencils: the pencil as the tool of the sketch, the temporary scratch of lines either for drawing or writing. While we work with a pencil, words or lines can be removed and replaced, and the draft is filled with arrows pointing to where chapters would change place or tiny letters where a whole sentence needs to be put into the finished text. The blue ink of the fountain pen demands structured writing, every stroke is final and each letter carries weight as to the presentation of the finished work. At school we were taught that errors had to be not scratched out but carefully lined out, using a ruler to make a straight line through our shameful error. This makes the drafting and the more or less careful calligraphy of the final version two distinctly different processes, in our minds as well as on paper.

253

254

Researching ICTs in Context

We are not positing that writing a weblog will change the articles we publish in scholarly journals. We do argue that blogging influences the way you think about thinking, and that it may change the process of research. To some extent it might even change the method.

Blogging software The word blog, now so popular, is commonly agreed to have been coined as late as in 1999, by Peter Merholz.14 That year also marks the start of the genre's speedy growth, which became more and more rapid as different tools made blogging simple. Blogger, Pita, Greymatter, Manila, Diaryland, Big Blog Tool and others15 offered free automated web publishing and often server space to go with it. Blogger has been the unequivocally most popular and most influential weblogging tool. Blogger was created by the company Pyra as a sidestep to the project they were really planning to develop, and became spectacularly popular, now boasting several hundred thousand users. Though Blogger initially failed to make a profit and had to lose all but one staff member in January 2001, the tool is still extremely popular and has recently been released in a for-pay version that offers additional features to the free system. Blogger has features that very much affect the way in which we work with weblogs.16 Blogger is set up to work over a web interface. The program and your writing are stored on a distant server. You log into blogger.com, write your post, press the publish button, and your words are uploaded to your website, automatically formatted and added to previous posts with the layout and style that you've set up. Being web-based, Blogger makes it easiest to write for your weblog while you're online. A 'blog this' button can be added to the toolbar of your browser, easing the connection between online reading and writing – if you click the button while viewing a web page, Blogger will automatically set up a writing space for you with a link to that page and space for you to write your comments. We started our weblogs using Blogger, and also use it for blogonblog.blogspot.com, the blog we have used in the process of researching and writing this article. Some months ago, Jill Walker stopped using Blogger and started using Tinderbox for writing and publishing her personal weblog.

Blogging thoughts

Figure 1: A recent example of our blogging. Tinderbox is an application designed primarily for writing, publishing and analysing notes, and it can also be used to publish your notes online, for instance as a weblog.17 Tinderbox lives on your personal computer rather than being web-based. It allows much more complex ways of writing and linking notes than Blogger does, but lacks the immediacy of Blogger's instant publication. In writing in Tinderbox there's a stronger inclination towards the private, the individual and enclosed space than there is in Blogger. This is partly due to the possibility of keeping some notes private, which is impossible in Blogger. Also the map views available in Tinderbox allow for spatial and colourful organisation that is intended for the writer, and that cannot easily be transferred to the web. This is a tool much more slanted towards the process of writing than towards the reader, though publishing can be done very elegantly with a little work from the writer. Blogger, on the other hand, resides on a server and its users can post new comments to their weblogs from any computer that has a web browser connected to

255

256

Researching ICTs in Context

the web. This offers a different kind of flexibility to Tinderbox, which limits you to your own computer. Methodology is not usually related to such quotidian details as the mechanics of the tools you use to write in, although sociology has been very aware of the potential of software in processing quantitative data. In the humanities, we are used to treating our technology as transparent. Discussions of technology tend to be about storage and viewing: the concern about old films which have become spontaneously combustible, the loss of old radio programs due to the re-use of the magnetic tapes they were recorded on, the freedom of replaying television programs when they can be taped to video, allowing researchers to subject them to relaxed and minute analysis. One of our main claims in this chapter is that writing and the way we express thoughts change when you use different tools. Though different blogging tools produce web pages that appear the same at first sight, using a new tool can have many unexpected effects, as Jill chronicled in her blog: 'I miss those quick little remarks', said Thomas over lunch. 'What happened?' Ah. I do write differently in Tinderbox. Something about the size of the writing space which invites more words. And something about the feel of it being a serious place where each post takes space on my screen. And something about the immediacy of Blogger where a button on my browser instantly lets me 'blog this' - I miss that button. I always have Tinderbox open so writing is quick but still not quite as 'chatty' as the 'blog this' button in my browser.18

Between the private and public This anecdote demonstrates the way in which weblogs straddle the boundaries between publication and process, between writing towards others and writing for oneself. A weblog is always both for oneself and for one's readers. If it were only for oneself, a private diary would be more useful. If it were only for readers, and not a tool for oneself, a more polished and finished form of publication would probably be more appropriate.19 Blogs exist right on this border between what's private and what's public, and often we see that they disappear deep into the private sphere and reveal far too much information about the writer. When a

Blogging thoughts

blog is good, it contains a tension between the two spheres, as delicate a balancing act as the conversation of any experienced guest of the French salons of the 19th century. The Theory of the Public Sphere is one of the most popular supports for the role of the news media as guardians of the democracy and the rights of the public to be informed. Jürgen Habermas20 did not claim, however, that the modern media are particularly well-adjusted to the ideal of a rational discourse in the public sphere. In his writings on the public sphere Habermas positions the public conversations geographically to the salons and the coffee-houses. All social classes were not included in this public, which consisted of the bourgeois: officials, scholars, pastors, with some overlapping to the still influential classes of craftsmen and small merchants: 'in the salons of the fashionable ladies, noble as well as bourgeois, sons of princes and counts associated with sons of watchmakers and shopkeepers.'21 The public sphere of political and cultural discourse exists between the private sphere of economy and the sphere of public authority, in a position of negotiation between the two: 'In this stratum, which more than any other was affected and called upon by mercantilist policies, the state authorities evoked a resonance leading the publicum, the abstract counterpart of public authority, into an awareness of itself as the latter's opponent, that is, as the public of the now emerging public sphere of civil society.'22 Habermas argues that the public sphere of the civil society has been colonised (or perhaps reclaimed) by a new nobility, politicians or stars known from the market of culture products who do not take part in the public sphere as much as present themselves in it. He calls this refeudalisation. This is supposedly the case today: we have a public arena where we do not participate, but acclaim the antics of the real actors. In this public of acclamation and performance, the real tension between the private and the public is gone: the private no longer has any real potential of influence, it has been made part of the public show: we know far too much about the sex-life of American presidents and other people with influence over our lives, and far too little about their thoughts, ideas and decisions, an argument Richard Sennet pursues in The Fall of Public Man.23 The salon existed on the borderline between the private and the public: it was situated in private homes, but part of the public sphere through being the site of the performance that was the salon-experience.

257

258

Researching ICTs in Context

The three aspects of Habermas's public sphere based on the salons or coffeehouses were: not equality, but a total disregard of status, the problematisation of areas that until then had not been questioned, and the principle of the public as inclusive, not exclusive.24

Public space expanded: personal publication Weblogs stand where the salon did: between private and public. A blog is written by an individual and expresses the attitude and the conviction of its writer; it is strictly subjective though not necessarily intimate. This doesn't stop it from being in the public domain, and being concerned about questions which are in the domain of public authority. Each individual can use weblogs as he or she feels fit, there is no tyranny of news values25 to decide what is worth writing about or, as the term is: what is worth blogging. This new medium of personal expression is another expansion of the public sphere into the private. When the news became concerned with the business ventures of citizens, that was an intrusion in what had been considered a private sphere, as we can still discern in the use of the expression 'that's his business', business being an arena into which we should not pry. But the intrusion was not forced, it was invited, just as the salons were privately owned rooms open to the public. With the weblog the public is invited into the privacy of the diary of an individual. This individual can seduce, attack, manipulate, rant or expose herself – but most of the time what you find in weblogs is an attempt to say something about what concerns the writer. In some manner, the writer is putting his or her daily experiences into a larger context, discussing micro events in relation to the wider universe of events. The weblog connects the public arena with that of individuals. As individuals' filters to the public sphere around them, weblogs express the personality and interests of their writers as well as the news. In this way blogs are a continuous way of writing oneself, as Daniel Chandler describes us writing ourselves through our traditional home pages.26 September 11th was an example of how personal opinion and experience needed a medium of communication that was not dependent on or obedient to the public authorities. With the aggressive stance of the US Government against all terror, defining terror as acts of

Blogging thoughts

aggression committed by Afghanistan Muslims, weblogs filled with shock and later discussion. The discussions in the weblogs were not in accordance with the authorities. Despite the univocal grief and shock at the plane-crashes, the online public did not blindly support the decision to attack Afghanistan or the changes to US legislation, just as they did not accept only one side of the story about what happened in Afghanistan. Online personal writers saw the war on Afghanistan and US military politics from many different perspectives.27 Instead of coming to a consensus opinion, as much of the traditional media did, weblogs showed the dissent among individuals. Bloggers were very vocal about their opinions and reactions to September 11th and the aftermath. On the day of the attacks, there were 22% more posts to Blogger users' weblogs than on an average day.28 In the weeks following the focus shifted from personal experiences and anxieties to opinions on the war. In addition to chronicling the regular news, many alternative news sources were linked to, and discussions raged between individual blogs and on community blogs like Metafilter. The myriad of blogging voices never accepted the news as a US Army promotional event as has been shown happened in the news coverage of the Gulf War in 1991.29

Autonomy & openness Weblogs are densely interlinked. This anchors blogs in the public arena, as part of a communal discourse. Posts to a blog can be very short and unpretentious. The threshold for publishing a single post is very low. This allows single, small, insignificant ideas to be expressed and formulated. Sometimes these thoughts are left as they are. A paragraph is enough and there is no more needed. Other times, the ideas grow. Someone links their site to the first post, comments on it, and a conversation grows forth. The initial post, or follow-ups, are linked to a web site or a newspaper article or something else. Links are like roots, tendrils, reaching out between fragments, creating a context for bits and pieces that at first glance may seem to be unconnected fragments. The individual weblog itself also creates a context for posts. Tom Matrullo compares blogs to loci amoeni: safe, idyllic, enclosed gardens where heroes of romance literature would recover from the battles of the outside world.

259

260

Researching ICTs in Context

The blog takes on some of the characteristics of the enclosed Renaissance garden, the interior plenitude of the autonomous voice reflecting upon and responding to other voices. (..) [I]f one goes back and looks into the worlds of folks like Ariosto, Tasso, Cervantes, etc., one begins to see that the gardens in their works are places of rebirth where the battered warriors briefly step outside the battle to re-collect themselves.30

This image encompasses the seemingly paradoxical mixture of private and public that is evident in weblogs. They are enclosed and private spaces that allow the writer to cultivate an autonomous voice. And yet they are visible, open spaces that encourage linking and conversations. Readers are welcome. Anyone can participate in the conversation simply by linking to the blog, though not everyone will be read. Unlike loci amoeni, blogs are not shut off from the world. They cannot be materially changed by references to them, but the references are visible to those who search for them, allowing each weblog to become part of a conversation with others. Our own blogs are safe spaces. Though you cannot stop others from flaming or criticising you outside your blog, within it you have total editorial control.

Academics and audiences Writing and publishing weblogs allows scholars to have a different relationship with their audience than we have with readers of traditional academic articles. Although journalists and scholars both write 'articles', the academic article is a far cry from the journalistic article. What journalists call research is scoffed at among scholars, and the many standards of the form of the scholarly article, such as the Chicago Manual of Style Form Guide, Humanities Version, describe a form developed for displaying the rituals of academic research and the equally ritualised formula of academic publishing.

Writing for our peers An academic article is predominantly written for other scholars: our peers. In the humanities and social sciences there are a few items it must contain: a) References to theory, preferably updated, focused, wide enough to display a healthy variety to our reading but not so wide that we can be accused of being shallow.

Blogging thoughts

b) References to empiric data, or the object of our research. This should have enough status that we can justify spending our valuable time on it, either through being a classic, being something entirely new, or being popular enough to have social significance. c) An original and spirited discussion of how a) relates to b). As long as our peers can see that we have met the demands of this formula, how we got to the point that our work and research can be presented in this manner is unimportant. Unless we are discussing methodology, how we actually made our discoveries is less important than the presentation of our findings, the discussion in relation to already existing theory and the defence of the hypothesis we derive from these results. The modern researcher is not quite the old man in a dusty, smoky study behind a labyrinth of books, unable to relate to the rest of the world. Research is supposed to be related to contemporary topics, and preferably lead to results that can give new insights to more than a narrow group of specialists. But the formula of academic writing is not designed to reach audiences different from ourselves; on the contrary, these ritualised forms limit readership to those who are rigorously trained to read a certain style of writing. With the increasing flow of and access to information, academic communication is changing. Through being aware of new communication technologies we academics can be part of the new writing and take part in the development of these emerging genres, instead of marginalizing ourselves. One defining characteristic of academic writing is the rigorous and formal citation practice. On the surface, weblogs seem like popular rather than academic non-fiction in that references are random and range from linking through written descriptions to casual mentions of sources. They however frequently refer as explicitly as do academic texts, though more simply by linking to a book's page at Amazon or to the web page referred to. Weblogs are written in order to share experiences rather than just display them, and for that the readers need to be able to find the books, music or web sites mentioned. Where academic writing is structured by the rules of the causal argument, a weblog is structured by time and the impulses of the day, documenting rather than structuring the trail of thought. The metaphors used in naming the sites: thinking with my fingers, Surftrail, Klastrup's Cataclysms, Jill/txt, all point to a structure not submitting to a academically accepted logic, but

261

262

Researching ICTs in Context

to a personal logic. This difference is much like the distinction between a connotation and an association, except that in weblog clusters the associations are shared through the blogs and become part of the culture, and thus take on the characteristics of a connotation. Ideas and concepts, or memes, can spread swift as viruses from blog to blog.31

Cultural Capital: the currency of scholars The rigorous training of an academic is basically rewarded through cultural capital. As Pierre Bourdieu has shown,32 cultural capital and economic capital can be mutually reinforcing, but are not prerequisites of each other. Academic criticism and public authority are also mutually reinforcing, leaving the scholar in a position of social status and influence, but at the same time impotent when it comes to offering criticism which can cause real system change.33 From the isolated splendour of ivory towers within the labyrinth of universities and research centres, one path to independence of a structure where even criticism of the system reinforces the power of the system is through breaking out of the pattern: to embrace the form which does not confirm the authority. As long as the energy of scholars is directed towards gaining status through the traditional channels of publishing and lecturing, the communication of scholarship will remain exclusive. To break the pattern, we need to think of new channels and new approaches to not only academic publishing, but to what merits publishing, at what point in the research process and how.34 The current reward system depends on certain formulas of academic publishing that encourage exclusivity and the fear of being robbed of thoughts and ideas. Since the real currency in the trade of academia is originality of thought and imaginative development of theories, there is more to lose than to gain in exposing your own ideas too early. The danger of having thoughts, ideas or questions copied before they have been published is not just a matter of some petty game between jealous professors with too little time on their hands, it's a very real matter of being robbed of the currency which measures academic success. From this point of view a weblog that reveals the thoughts, arguments and questions of the scholar continuously during the process of research and long before academically accepted publication in print seems like a waste of perfectly good imagination and theory develop-

Blogging thoughts

ment, an invitation to having your ideas looted. On the other hand, published and archived in the World Wide Web, the same ideas and thoughts are in fact published and as such better protected than if they were for instance given away over a cup of coffee, randomly at a conference. As we were finishing this paper, a brief article on weblogs, 'Blog This', was published by Henry Jenkins, noted media scholar at MIT. This is the first article published on weblogs by an academic, as far as we have been able to ascertain. The thesis of Jenkins' article is that weblogs are powerful and may become even more influential by filling the current lull in commercialisation of the Internet. However, the words that he uses clearly show that he thinks of bloggers as a very different species to himself, and presumably, to the other academics he is primarily writing for: Like cockroaches after nuclear war, online diarists rule an Internet strewn with failed dot coms. (…) Bloggers are turning the hunting and gathering, sampling and critiquing the rest of us do online into an extreme sport. We surf the Web; these guys snowboard it. Bloggers are the minutemen of the digital revolution.35

In the days after the article was published, bloggers have indeed 'blogged this', as Jenkins invited them to, bringing Jenkins' article to second place in the 'most linked to' URLs at Blogdex on 16 February 2002. Dave Winer wrote one of the earliest and most extensive commentaries on the article. He was clearly offended at being called a cockroach, and at Jenkins' obvious amazement at the power of bloggers. Winer defends himself and us other bloggers by angrily pointing out that in the late 70s, IBM didn't think the personal computing community was a threat either, and see where that led!36 Jenkins would probably agree with Winer, but his style alienates the bloggers he praises. In his use of language, Jenkins treats bloggers as objects of research; and to some extent, as objects of wonder to be exhibited as extreme and freakish ('cockroaches', 'extreme sports'). This objectification and alienation is in stark contrast to bloggers' own perception of their community as made up of writing, verbal, influential subjects. The distancing techniques Jenkins uses, and his unfortunate choice of imagery, are typical for academics, who are so used to studying new technologies as exotic objects that they fail to see that they could be useful within academia itself. It is interesting that the pivotal event which made Jenkins write of weblogs, took place at a conference where weblogs were an important part of the conference publications.37

263

264

Researching ICTs in Context

A few months ago, I was at the Camden Pop!Tech conference, and the guy sitting next to me was typing incessantly into his wireless laptop, making notes on the speakers, finding relevant links and then hitting the send key— instantly updating his Web site. No sooner did he do so than he would get responses back from readers around the country. He was a blogger.38

The speed with which the comments on keynotes and speeches were published, defies the control of peer-reviewed publications and other authoritative academic voices. Perhaps Henry Jenkins is worried that he might lose control. That would be a feasible fear for a senior and well respected scholar faced with the rapid changes of online discourse. In his article he even tries to take control of the responses of the people blogging his article, through the last couple of paragraphs where he firstly tells us all what different reactions we will respond with, and secondly asks if we will please blog him. The cockroach reference in the quote was interestingly enough quickly altered after Jenkins' article Blog This had been blogged. Webloggers' offended reaction to the introduction made Jenkins request that his editor remove the offending phrase. He also emailed his students, describing his feelings about the reactions to the article, and permitted Elin Sjursen, an MIT graduate student and the writer of the blog BloggerdyDoc, to publish these emails in her blog.39 Engaging in a discussion about blogs, Jenkins needed a public voice on the web in order to defend his position. Without an online platform to speak from, the traditional relationship of researcher to object became inverted as the bloggers he had written about started analysing their analyst. Trying to sustain the academic distance built into our traditional scholarly or popularising forms of publication can result in a greater loss of control than participating fully in these new forms of communication and conversation.

Thinking with computers Computers were deliberately designed to reflect and augment our thinking. Vannevar Bush, a prominent developer of analogue computers, argued for mechanical, non-hierarchical ways of organising information which would be more suited to the associative thought patterns of our brains. In a 1949 article tellingly titled 'As We May Think', Bush sketches designs for a device he called the memex. Though never realised, Bush's descriptions and thoughts about the memex are commonly

Blogging thoughts

seen as direct ancestors of today's digital hypertext. Later, Doug Engelbart and his team developed the graphical user interface, the mouse and other elements that we take for granted in today's computer interfaces with the explicit aim of augmenting the human intellect . And yet technology is always unpredictable. Computers have changed our lives in many ways unforeseen by pioneers such as Bush and Engelbart. In the last decade, sceptics have warned against the fragmentation of digital media. Supposedly, our attention spans are decreasing dramatically and we are losing our capacity for sustained reflection.40 In some ways, weblogs can be seen to be a genre that reflects these concerns, confirming the sceptics' worst fears. To examine how weblogs work upon us it is necessary to scrutinise both the shape of this new genre and the ways in which the form encourages us to write and connect information.

Writing & thinking in weblogs Weblogs are collections of briefly formulated thoughts and ideas, very much in contrast to the lengthy, sustained argumentation we expect of a scholarly article. Instead weblogs focus on connections and on brief nuggets of thought. Links are vital to the genre; take the links out of a weblog and you are left with a web diary, a much more introverted and private form of writing. A blogger can be seen as a modern version of Vannevar Bush's trail blazers: a person who links separate documents together, creating a trail or a path through them for others to follow. There is a new profession of trail blazers, those who find delight in the task of establishing useful trails through the enormous mass of the common record. The inheritance from the master becomes, not only his additions to the world's record, but for his disciples the entire scaffolding by which they were erected.41

This task of connecting information that is already available is part of the work of research as well. For researchers who are studying online phenomenons, the weblog is perfectly suited to this work of connecting dispersed discoveries, at the same time as a weblog allows us to share this found information, and to participate in discussions about it. Trailblazing in a weblog can be an element of research and a dissemination of that research at the same time.

265

266

Researching ICTs in Context

Weblogs link to external sources as well as to other weblogs. Extensive discussions take place between blogs, with links referring to previous comments in the conversation. These dialogues, or perhaps better, polylogues,42 can seem complicated in that the structure is associative and idiosyncratic rather than hierarchical and externally ordered, and yet the constant links between the blogs make the discussion easy to follow – and more interestingly, easy to continue. We post to our blogs as ideas come to us. Daily, hourly, weekly; the frequency varies but it is a writing that happens in bits and pieces, not in the long hours of thought that suit the clichéd image of the secluded scholar in the ivory tower. In this sense blogs are suited to the short attention span of our time that worries so many traditionalists.43 Blogs are interstitial for the writer as for the reader. A post in a blog will often form a chain of thought with other posts and other fragments, but can stand alone as well. Weblogs are written continuously and published without being revised. Though a tool like Blogger does allow a post to be written, saved and not published instantly, publishing a new post will automatically also publish the drafted post. The system assumes instant publication will be the norm. Sometimes webloggers will revise posts later and republish them; other bloggers make it a matter of principle to limit revision, preferring the immediacy and perhaps, in a sense, honesty, of the first expression of a thought. The tagline for Jouke Kleezenbaum's weblog Notes, Quotes, Provocations and Other Fair Use, for instance, is 'the mark of launch-andlearn publishing: corrections are generally made within 36 hours'. This instant publication encourages spontaneous writing rather than carefully thought out arguments. Being allowed to write spontaneously releases us of the expectation that our writing must be perfect and polished. While most weblogs are personal and informal in their purpose, academics writing in this genre also use a much more informal writing style than they tend to in articles written for publication. Adrian Miles comments on this in his weblog: What is novel is the tone that I'm adopting. Ordinarily when I write in any text app that is not visibly online (email, html or MOOs) my tone is quite different. Authorial, authoritative, academic, scholarly, teacherly (yeah, right), and various other 'formal' modes of utterance. Blogs have changed that dramatically with a much more informal tone but with some of the hallmarks of print literacy. It resides in the public domain, they are authorial to the extent that they are written and published, and they career

Blogging thoughts

the imprimatur of semi permanence and care that email and other more 'oral' forms of electronic literacy don't quite get to.44

In our blogs, we allow ourselves to write half-thought, naked ideas and show them to others rather than saving them for fully fleshed out carefully thought through papers. Most ideas a researcher has never make it to a formal article or book; they are forgotten or cast aside because they don't fit the whole. Weblogs elucidate the constant flow of thought and the ever-changing nature of research. Blogs are chronologically ordered, rather than ordered according to the logic of an argument or the persuasive patterns of rhetoric. Their order is determined by the time of thinking. In contrast to the logical and topical organisation usual in academic writing and note-taking, blogs are chronologically ordered. Writing in a medium – or perhaps better, an element – which encourages a different way of structuring thought can enable us to see differently. Roland Barthes is a writer who deliberately attempted to escape the tyranny of logical argumentation by arranging his writing alphabetically, sectioning his book into segments arranged by the order of their titles. The alphabetical order erases everything, banishes every origin. Perhaps in places, certain fragments seem to follow one another by some affinity, but the important thing is that these little networks not be connected, that they not slide into a single enormous network which would be the structure of the book, its meaning. It is in order to halt, to deflect, to divide this descent of discourse toward a destiny of the subject, that at certain moments the alphabet calls you to order (to disorder) and says: Cut! Resume the story in another way.45

The chronological order of weblogs is from one perspective less random than the alphabetical order Barthes used. Posts to a weblog are published in the order they were written. Thought follows thought and one idea tends to trigger another, though not necessarily in a causal chain. On the other hand, it can be argued that the order of weblogs is more random than Barthes' alphabetically ordered books. Alphabetisation is a structure edited and worked through and the author organises and names each fragment to create a whole, finished work to be published all at once. Weblogs are published bit by bit; they are always in progress, always becoming. Unedited, spontaneous, scrolling away so the most recent thoughts are always at the top of the page and the older ideas are harder to get to,

267

268

Researching ICTs in Context

blogs encourage a feeling of time which is very different to that felt in traditional academic writing. Usually, notes written on paper accrete in piles. We group them by topic rather than by when they were written. Articles are drafted, revised, rewritten, edited, proofread, finalised and the notes forgotten, made irrelevant by the more polished product. Blogs are a way to trace the flight of thought rather than the chain of thought. The tangible work of research in the humanities is reading, thinking and writing. In blogs these collapse into one movement. Blogs assume linking and reading and are the written trace of these activities. A blog is a trail, a visible trace of the process of research. Writing a blog presupposes that you read – and all the blogs we've seen refer to other blogs. When we read on the web, our reading is affected by the technology we use to access the text: computer hardware, operating system, browser. Imagine the frustration of reading a piece written as a hypertext on a computer printout with no chance to follow links. To summarise: blogging encourages spontaneous, timely and concise expression of thoughts. The genre has in it an expectation of linking, so posts will often comment on other writing and be linked to it.

Archiving and analysing Weblogs are used for recording thoughts, for sharing thoughts, for participating in discussions, and also for analysing thoughts. Rebecca Blood, whose article on the history of blogging has received much acclaim in the blogging community, expresses this in these words: Shortly after I began producing Rebecca's Pocket I noticed two side effects I had not expected. First, I discovered my own interests. I thought I knew what I was interested in, but after linking stories for a few months I could see that I was much more interested in science, archeology, and issues of injustice than I had realized. More importantly, I began to value more highly my own point of view. In composing my link text every day I carefully considered my own opinions and ideas, and I began to feel that my perspective was unique and important.46

This experience is echoed by many bloggers. From a junior scholar's point of view, blogging can be an excellent method for developing and sustaining a confident and clear voice of one's own and the ability to formulate and stand by opinions. While private journals may fill with notes, they need not be as clearly formulated as a post in a weblog,

Blogging thoughts

which is intended to be read by others, and it is easy to neglect old notes scribbled in the margin of a book or on a notepad. Writing in a weblog one is forced to confront one's own writing and opinions and to see them reflected in the words of others. The discussions are much more open and also more permanent than discussions in a seminar room or at a conference. A blog is a permanent archive (as long as the writer preserves the archives and the server remains online) and it is searchable. What you write in your blog can be quoted and discussed in any forum. When we write in a blog thoughts don't disappear. They stay. A thought we might otherwise not care to record or mention can be what triggers new work. An example is this passage from thinking with my fingers: I also have selfish reasons for blogging. I think better when I write. Sometimes, I need to get rid of thoughts, and then I write them down so that I can go on. When I was 16 I wrote down the names of the boys I was in love with. If there was one I happened to hate I would burn the note afterwards, and feel like I had some kind of closure. Now, when I am in love with a thought, I can write it down. That lets me examine it when it doesn't expect me to look at it. I can sneak up on it at a time when my head is busy with something else, and I can surprise it in a different context. This will let me see my newfound love, the virgin idea, in a different light, and I can see its flaws and weaknesses, as well as its beauty. And I can move on, let the ones which are not worthy of being taken home live on somewhere outside my head.47

This post, a scrap of memory from the seventies combined with the experience of writing the blog itself, stayed with Jill, the other author of this paper, for long enough to resonate and set new thoughts into flight, ultimately leading to this paper. Today it's possible to go back into the archive and find that scrap which otherwise most likely would not have been written down and certainly not published.

Subjective and concrete Because weblog posts are usually written in relation to another text or to an experience, they tend to be concrete in nature. Sometime the concrete beginning leads to an abstraction or to a theoretical thought. Abstract and objective thought constitute knowledge in the paradigm of literature cultures; this stands in contrast to what is perceived as knowledge in an oral culture. In a discussion of Walter Ong's work on orality and literacy, Michael Heim writes that

269

270

Researching ICTs in Context

[R]ealities in oral cultures are preserved by metrical language with repeatable rhythmic formulas. Since cognitive form requires patterned, rhythmic speech, what is knowable in those cultures is what falls neatly into the commonplaces and formulas. What in oral cultures cannot be fitted into the topics or stable formulas much be disregarded as arbitrary, absurd, or at least of lesser significance.48

How does this relate to blogs? Blogs could be seen to have certain, oftenreturned-to topoi: the remembering of past experiences and relating them to some gained insight, the recognition and marking of respect when a blogger links to someone else who's said something that the writer agrees with or has experienced or has tried to express, the linkslutting,49 the memes, the meta-reflection about the nature of blogging, linking and popularity, the obsession with recognition and people linking to me, the musings about why I write or why I have decided not to write. These are in a sense the formulae of blogs. The speech itself is not patterned and rhythmic, though it does generally have an informal tone that could perhaps be differentiated from the styles common in journalism, academia, technical writing or fiction. The topoi are recurrent motifs of blogs though. In addition to the words, the visual appearance of weblogs is important. There are standard templates that are used, unaltered, by many bloggers. Others adjust these layouts or create their own from scratch. Even the most individual layouts are made up of elements chosen from the same sack. The page is almost always divided into two or three columns, where the narrower left column includes a description of the weblog, links to previous months' archived posts, links to other blogs that the blogger reads. Sometimes some of these points are in the right hand column instead; sometimes the columns are reversed. Posts are usually set apart in some way: they may be time stamped, dated, signed («posted by NN»), titled or simply separated by a line or blank space. Many weblogs participate in web rings and display colourful banners or buttons which both declare their allegiance and allow the reader to click through to another, presumably somehow similar, weblog. Some weblogs have no or very few images while others rely on rich eye candy. The manga-inspired design of many sites is an example of how not only ideas (memes) but also images and design elements spread fast through the network of weblogs. Traditional academic essays also follow certain topics and formula, as mentioned above. The topoi of weblogs, visual, verbal and conceptual, do appear to be more virulently repetitive. It would be worth investigating

Blogging thoughts

whether the memes so often referred to among bloggers are accepted and spread precisely because they do fit neatly into the topoi – the commonplaces and formulae – of weblogs. The idea that some kinds of knowledge are discarded as absurd because they do not fit into the stable formulas of a genre could be applied to both weblogs and to the scholarly article or book. This is an important reason for writing in different ways, with different tools and in different genres, so we can appreciate many different kinds of knowledge.

The social network of blogging Weblogs are written in relation to other texts. As discussed above, the primacy of the link affects the style of writing; it affects the form or the genre. It is often stated that it is hard to define a weblog,50 yet as we have shown, there are simple formal qualities common to all the sites that call themselves weblogs. Evan William's succinct 'Frequency, Brevity, and Personality' provides a good summary of these formal traits. Within the blogging community, there is less agreement about what a good weblog is like. In such a group of determined and vocal individuals, where the number of weblogs is far too great51 for any one reader to have sampled all or even most of the blogs in existence, quality is not a satisfactory or practical measure. Popularity, on the other hand, is easily measurable if you interpret popularity as the number of other weblogs linking to a particular blog. Numbers of readers are rarely mentioned, and these statistics are in any case untrustworthy and not publicly verifiable. Links to a site can be found by anyone, by using a search engines or by looking at a site like Blogdex, which indexes links between weblogs. Weblogs tend to come together in clusters as they link to each other. A reader of your site may link to you; you see the link in your referral stats and start reading their blog. You find it interesting, and link back to it. The readers of your blog, some of who keep their own blogs, start reading the other blog, and some of them also link to it. And so it continues. Our weblogs belong to such a loose cluster, as before mentioned, 'the Scandinavian flavoured cluster'52. In part the cluster has grown forth as described here, through gradual linking. It has also developed through personal and professional contacts; many of us knew each other from conferences before starting to write weblogs. In addition, we work within adjacent fields: game studies, hypertext, online narratives and so on.

271

272

Researching ICTs in Context

In the wider blogging community the most popular, wellestablished and famous cluster is known as the 'A-list'. The A-list is a cluster of early adaptors to weblog technology, some of them creators of popular software, such as Evan Williams, co-creator of Blogger, or people who are part of the same (offline) social network as the creators. These extremely popular blogs are not necessarily distinguished through style, but through persistence: writing a fairly interesting weblog over time increases the chance of being read, linked to and 'blogged' by others. Because the A-list is linked to fame through persistence and in some cases early presentation, rather than quality, it is static and unchangeable. Cameron Marlow, the developer of Blogdex, is currently working on a PhD project at the MIT Media Lab where he is using data on linking structures between blogs to analyse what he refers to as the social networks between blogs and the relation between this network and the ways in which memes spread between blogs. He has developed the Blogdex indexing system as part of this research. The Social Network Explorer uses information from Blogdex's database to show the networks between blogs. While references in academic papers (which are in many ways close to hypertextual links) are rarely thought of as markers of friendship, Marlow's Social Network Explorer uses the term 'friends' for the persons writing the weblogs with the closest connections through linking - the most frequent links between you and it. To be told by a program that you are the friend of a person you have never met, can be constructed as intrusive - but knowing that the members of the A-list, the early users of weblog software, used the web not to reach a large audience but to reach each other; to talk of two people with several links back and forth between their sites as 'friends' becomes logical and perhaps even precise. It also demonstrates the highly social nature of weblogging. In the semi-social structure of weblogs, linking is a measure of popularity, which again is generally interpreted as a measure of quality. The poetics of weblogs is not present in these measures of quality, and might seem to be non-existent. But the poetics and aesthetics of the form is a central topic of weblogs themselves: an endless meta-discussion in between the other topics of choice, a common denominator and source of conversations.

Blogging thoughts

Conclusions This entire article might be dismissed as an elaborate excuse for procrastinating while we should be finishing our PhDs. Maintaining a weblog while we complain about the stress of writing a dissertation seems self-contradictory at first and second glance. Both of us however experienced that writing the thesis became easier and the writing more focused after we started blogging. This discord between commonly accepted ways of studying for and writing a thesis and our own experiences was the motivation for this discussion. Our positive experiences with blogs may be related to the fact that we both do online research. Rather than distancing ourselves and permitting an escape from the object of research, the blog lives within the same frame as the computer games and the electronic narratives we study, keeping us close to the technology, the relevant formal as well as informal discourse and the objects themselves. The weblog as an equalising power on the net and an equaliser between scholars and non-scholars is another matter all together. While services like Blogger in principle make it possible for everybody to start their own blog, participate in discussions and be part of the blogging community, the different clusters grow slowly, expand conservatively and put great value in exclusivity expressed through links. To be able to prove this claim, we would need to expand the scope of our study beyond our own cluster. This is the curse of studying cutting-edge technology – to study it you have to be an innovator or early user, and as such your experiences and your knowledge cannot be immediately generalised. All we can do is to speak for ourselves and present our experiences. We open our claims to scrutiny, point to sources that might assist or deny our claims and offer this article to our peers, in the best tradition of academia.

References Adler, C. 2002. 'Weblogs by pros are founts of insight'. Fortune.com [online, cited 12 February 2002]. At: http://www.fortune.com/indexw.jhtml?channel=artcol.jhtml&doc_id=206170 Barrett, C. 1999. 'Anatomy of a Weblog'. CamWorld. 26 January 1999. [online, cited 14 February 2002]. At: http://www.camworld.com/journal/rants/99/01/26.html

273

274

Researching ICTs in Context

Barthes, R. 1975/1989. Roland Barthes by Roland Barthes. Ithaca, NY: Noonday Press. Bernstein, M. 2002. markbernstein.org. Weblog [online, cited 12 February 2002]. At: http://markbernstein.org Birkerts, S. 1994/1996. The Gutenberg Elegies: The Fate of Reading in an Electronic Age. London: Faber and Faber. Blogdex. 2001-2002. 'Weblog indexing system'. [online] Cambridge MA: MIT Media Lab. [cited 18 February 2002]. At: http://blogdex.media.mit.edu Blood, R. 2000. 'Weblogs: a history and perspective.' Rebecca's Pocket. 7/9/2000 [online, cited 9 February 2002.] At: http://www.rebeccablood.net/essays/weblog_history.html Blood, R. Forthcoming, June 2002. We've Got Blog: How Weblogs are Changing our Culture. Cambridge, MA: Perseus Publishing. Bourdieu, P. 1984. Distinction, a Social Critique of the Judgement of Taste. London: Routledge. Bourdieu, P. & Passeron. J.-C. 1990. Reproduction in Education, Society and Culture. 2nd ed. London: Sage. Bush, V. 1945. 'As we may think.' Atlantic Monthly, Vol. 176, No. 1: 85-110. [online edition also available, cited 18 February 2002]. At: http://www.theatlantic.com/unbound/flashbks/computer/bushf.htm Chandler, D. 1998. 'Personal home pages and the construction of identities on the web'. [online, cited 14 February 2002.] At: http://www.aber.ac.uk/media/Documents/short/webident.html Chicago Manual of Style Form Guide. [online, cited 8 February 2002] At: http://www.lib.ohio-state.edu/guides/chicagogd.html Corneliussen, H. 2001-2002. Gender and Computing. Weblog. [online, cited 12 February 2002]. At: http://hildesin.blogspot.com Dvorak, J. C. 2002. 'The Blog Phenomenon'. PC Magazine [online, cited 12 February 2002]. At: http://www.pcmag.com/article/0,2997,s=1500&a=21865,00.asp Engelbart, D.C. (1962). 'Augmenting human intellect: a conceptual framework'. California: Stanford Research Institute. [online edition also available, cited 12 February 2002]. At: http://www.histech.rwthaachen.de/www/quellen/engelbart/ahi62index.html Fagerjord, A. 2002. Surftrail. Weblog. [Cited 12 February 2001]. At: http://www.media.uio.no/personer/andersf/blog/

Blogging thoughts

Galtung, J. & Ruge, M.H. (1965). 'The structure of foreign news.' Journal of Peace Research, 2. 65 – 91. Habermas, J. 1989. The Structural Transformation of the Public Sphere. An inquiry into a category of bourgeois society. Cambridge: Polity Press. Heim, M. 1999. Electric Language. A philosophical study of word processing. 2nd Ed. New Haven: Yale UP. Jarvis, J. 2001-2002. Warlog: World War III. War, Media and the Web. [online, cited 10 February 2002]. At: http://www.buzzmachine.com/ Jenkins, H. 2002. 'Blog this: digital renaissance.' Technology Review: Emerging Technologies and Their Impact. [online]. Cambridge, MA: MIT. March. [cited 17 February 2002]. At: http://www.techreview.com Klastrup, L. 2000-2002. Klastrup's Cataclysms. Weblog. [online, cited 13 February 2002]. At: http://www.itu.dk/people/klastrup/ Kleerebezem, J. 2000-2002. Notes, Quotes, Provocations and other Fair Use. Weblog. [online, cited 9 February 2002]. At: http://nqpaofu.com/ Marlow, C. 2001-2002. Blogdex. Software and project notes. [online] Cambridge MA: MIT Media Lab. [cited 17 February 2002]. At: http://blogdex.media.mit.edu Matrullo, T. 2002. 'Loci Amoeni.' Commonplaces, 23 January 2002. Weblog. [online, cited 10 February 2002]. At: http://tom.weblogs.com Mead, R. 2002. 'You Got Blog!' NY Times, 13 November 2000. [online edition also available, cited 18 February 2002]. At: http://www.rebeccamead.com/2000_11_13_art_blog.htm Merholz, P. 1998-2002. Peterme. Weblog. [online, cited 17 February 2002]. At: http://peterme.com Miles, A. 2001-2002. Vlog. Weblog. [online, cited 14 February 2002]. At: http://hypertext.rmit.edu.au/vog/vlog Mortensen, T. 2001-2002. Thinking with my fingers. Weblog. [online, cited 9 February 2002]. At: http://torillsin.blogspot.com Mortensen, T. & Walker, J. 2001-2002. blogonblog. Weblog. [online, cited 15 February 2002]. At: http://blogonblog.blogspot.com Ottosen, R. 1991. The Gulf War with the Media as Hostage. Oslo: International Peace Research Institute. Pop!Tech 2001. Online, Everywhere, All the Time. How Technology will Change our Lives. Conference web site with archived weblogs. [online, cited 17 February 2002]. At: http://www.camcon.org/CTC_2001/Public/home.cfm

275

276

Researching ICTs in Context

Sjursen, E. 2001-2002. BloggerdyDoc. Weblog. [online, cited 11 March 2002]. At: http://www.student.uib.no/~stud2066/blogger/ Stærk, B. 2001-2002. The World After WTC. Weblog [online, cited 10 February 2002]. At: http://www.bearstrong.net/blog1.2 Sennet, R. 1977. The Fall of Public Man. New York: Knoppf. Stone, B. 2002. Weblog: The New Home Page. Weblog [cited 13 February 2002]. At: http://www.bizstone.com/book/ Walker, J. 2000-2002. jill/txt. Weblog. [online, cited 9 February 2002]. At: http://cmc.uib.no/jill Weblogg-Ed. 2001-2002. Weblog. [online, cited 13 February 2002]. At: http://www.ravenrock.com/blog/blog.html Weblogs in Education. 2002. Weblog. [online, cited 12 February 2002]. At: http://www.schoolblogs.com/ Weinberger, D. 2002. JOHO the Blog. Weblog. [online, cited 10 February 2002]. At: http://www.hyperorg.com Winer, D. 1998-2002. Scripting News. Weblog. [online, cited 16 February 2002]. At: http://scripting.com

Notes 1

Giles Turnbull.'The State of the Blog.' Interview with Even Williams, Writing the Web. 28 February 2001. 2 Mark Bernstein, markbernstein.org, 2002. 3 Mark Bernstein, Mark Bernstein, 1 September 2001: Someone should take a good look at clustering phenomena in Web logs. For example, consider the interesting and active cluster of media-theory Scandinavian-flavored weblogs from Sjursen, Walker, Klastrup, Frasca, Miles, and others. The Web design community has recently been sharply critical of co-citation practices, which some regard as merely a way for elites to reinforce their influence and which has led to the suspension of dreamless.org and K10K pending the arrival of cooler heads. I suspect, though, that these clusters are more interesting, and less strictly political, than they may seem at first — perhaps a visible manifestation of discipline-formation in process. 4 Anders Fagerjord, Surftrail, 2002. 5 Lisbeth Klastrup, Klastrups Cataclysms. 6 Anja Rau, Flickwerk. 7 Rebecca Blood, 'Weblogs: a history and perspective,' Rebecca's Pocket, 7/9/2000

Blogging thoughts 8

Reason P. & Bradbury, H. 2001. 'Introduction: inquiry and participation in search of a world worthy of human aspiration'. Handbook of Action Research. London: Sage. 9 Reason, 1. 10 Biz Stone is planning a book on weblogs and has started writing a blog about the process: http://www.bizstone.com/book/, in Fortune.com Carlyle Adler writes about professional weblogs, Weblogs by pros are founts of insight. In February 2002 John C. Dvorak in PC Magazine writes about blogs as a recent change to online writing in his article The Blog Phenomenon, but he does not consider professional blogs. 11 Weblogs are by definition always works in progress. The form defies finality. What is posted today may be withdrawn tomorrow. 12 Peter Merholz's Peterme and Mark Bernstein's Mark Bernstein are excellent examples of professional rather than university-based research blogs. 13 Heim, M. 1999. The Electric Word. New Haven: Yale UP. 32. 14 Peter Merholz, Peterme, 28 May 1999. 15 Blogdex has a longer list of different weblogging tools, both servers and clients: http://blogdex.media.mit.edu/info.asp. Cited 12 February 2002. 16 It is difficult to discuss weblogs without discussing the tools used to write them. Different software allows different processes, and the differences between software products can be subtle, but always control their users. 17 Tinderbox was developed by Eastgate Systems and released in 2002. More information on this product can be found at http://eastgate.com 18 Jill Walker, jill/txt, 20 November 2001. 19 On the other hand, it is possible that the raw nature of weblog writing is part of the reason for it's popularity. The immediacy, the continuity, the apparent honesty and the unedited quality of weblogs are traits comparable to important elements in reality TV shows. 20 Habermas, J. 1989. The Structural Transformation of the Public Sphere. An Inquiry into a Category of Bourgeois Society. Polity Press: 21 Habermas (1989: 33). 22 Habermas (1989: 23, original italics). 23 Sennet, R. 1977. The Fall of Public Man. New York: Knoppf. 24 Habermas (1989: 36). 25 Galtung, G. & Holmboe Ruge, M. (1965) The Structure of Foreign News. Journal of Peace Research Bd. 2). 65 – 91. 26 David Chandler, 'Personal home pages and the construction of identities on the web', 1998. 27 In the weeks after September 11 already existing weblogs such as Adnan Arif's adnan.org, Rebecca Blood's Rebecca's Pocket and many others became almost exclusively devoted to discussing the attacks and the war. Other we-

277

278

Researching ICTs in Context

blogs were started specifically to discuss the events, see for instance Jeff Jarvis's Warlog or Bjørn Stærk's The World after WTC. Blogger.com's archives for September 2001 contain links to newspaper coverage of webloggers writing about the attacks, and to individual weblogs. 28 Evan Williams posted this information to the front page of Blogger.com, 12 September 2001. 29 Ottosen, R. 1991. The Gulf War with the Media as Hostage. Oslo : International Peace Research Institute. 30 Tom Matrullo, 'Loci amoeni', Commonplaces, 23 January 2002. 31 The term meme, meaning a unit of cultural information that is transmitted from one mind to another, was coined by Richard Dawkins as a cultural equivalent to the gene in his book The Selfish Gene (Oxford: Oxford UP 1989). The term is heavily used among bloggers, giving a name to some blogs such as memepool.com and mememachine.net. 32 Bourdieu, P. 1984. Distinction, a Social Critique of the Judgement of Taste. London: Routledge. 39. 33 Bourdieu, P. & Jean-Claude Passeron, J.-C. 1997/1990. Reproduction in Education, Society and Culture. London: Sage. 31. 34 According to Bourdieu, the system will embrace rebellion quicker than you can produce new ways to undermine it. This paper can be seen as such an embrace: incorporating back into the sphere of scholarship the tool which lets us publish independently of it. 35 Jenkins, H. 2002, 'Blog this: digital renaissance', Technology Review (Cambridge MA: MIT). March. The first part of the initial sentence was removed from the online version of the article a few days after we cited it. 36 Dave Winer, Scripting News, 15 February 2002. 37 Pop!Tech 2001. Online, Everywhere, All the Time. How Technology will Change our Lives. The conference web site has a link to weblogs that were written during the conference containing notes from the keynotes and other observations from the conference. Readers could enter their comments in a discussion feature. 38 Jenkins, op.cit.. 39 Elin Sjursen, BloggerdyDoc, 28 February 2002. 40 See for instance Birkerts, S. 1996. The Gutenberg Elegies. London: Faber & Faber. 27. 41 Vannevar Bush, 'As we may think'. 42 David Weinberger refers to these polylogues as blogyarns in contrast to the commonly used blogthreads, because several blogs participate; each blog becomes a thread in the spun yarn. JOHO the Blog, 15 February 2002. 43 Birkerts, op.cit. 12. 44 Adrian Miles, vlog, 18 September 2001.

Blogging thoughts 45

Barthes, R. 1989. Roland Barthes by Roland Barthes. Ithaca, NY: Noonday Press. 148. 46 Rebecca Blood, 'Weblogs: a history and perspective.' Rebecca's Pocket. 7/9/2000. Blood also has a book on weblogs forthcoming from Perseus Publishing in June 2002. 47 Torill Mortensen, Thinking With My Fingers, 20 June 2001. 48 Heim, op.cit. 61. 49 Linkslutting is a term used for trying to get people to link to your blog in return for your linking to them, where the actual link (which is hoped to increase traffic to the site; i.e. the number of readers) is more important than any content. 50 See for instance The Daily Dose, 31 August 2001. At: http://www.thedailydose.net/dose310801p4.htm 51 Cameron Marlow estimates the total number of weblogs to be close to 400000. There are nearly 15000 indexed by Blogdex at the time of writing. (post to the Blogdex weblog 14 January 2002. At: http://blogdex.media.mit.edu/news/archives/00000071.asp 52 See note 3.

279