The European Network for Intelligent Information Interfaces

June 2002

magazine

12

Homo Ludens Computing and creativity

Technology on the right side of the brain Collaboration between artists and technologists Ambient Intelligence Ghostwriter Extreme interfaces

The European Network for Intelligent Information Interfaces

1

Editorial Mimo Caenepeel Designing for Homo Ludens (Feature) 2 Bill Gaver

Putting the art in artificial (Feature)

7

Geraint Wiggins

The art of collaboration (Feature)

12

Linda Candy

Creating Aesthetically Resonant Environments (i3 legacy)

17

Phil Ellis,Tony Brooks

Darwin and creationism reconciled (Feature)

20

ontents

magazine

Amilcar Cardoso, Carlos Bento, Perousal Machado, Francisco C. Pereira

Mon Meilleur AmI (Feature)

26

Ken Ducatel, Marc Bogdanowicz, Fabiana Scapolo, Jos leijten, Jean-Claude Burgelman

Computer games — for better or for worse? (Feature)

32

Judy Robertson

Extreme Interfaces (Report)

35

Patrick Purcell

Remembering Mike Scaife

38

Yvonne Rogers

News

39

Des visionnaires dans le temple du business (Report)

42

Mimo Caenepeel

The ethics of design after September 11 (Column)

45

Giorgio De Michelis

Networked Learning (Book review)

46

John Siraj-Blatchford

Future events

47 48

i3net member sites

Summer issue: Technology on the right side of the brain Mimo Caenepeel University of Edinburgh [email protected]

Could you imagine sharing your home with a computational alien, or using a digital boudoir? Perhaps you would find it easier to envisage becoming a meditative voyeur, or playing a competitive game in which the most relaxed person wins? Technology acquires a new dimension when it leaves the sphere of the workplace, with its emphasis on efficiency, productivity, and “getting things done” (important though that may be). Entering neighbourhoods, artist’s studios, lonely mountains tops…, technology has recently started to support different values and different kinds of pursuits. Authors in this issue mention aimlessness, wonder, curiosity, playfulness, and – many times – creativity. Bill Gaver conjures up new types of technology designed for “Homo Ludens”, after Huizinga’s wellknown definition of humans as playful creatures. Geraint Wiggins shows how the study of creativity can be approached from different perspectives, and introduces the “artificial artist”, AARON. AARON also crops up in Linda Candy’s article, which focuses on creative collaboration between artists and technologists. And Amilcar Cardoso and his colleagues in the Creative Systems Group sketch computer models of creativity, which may surprise by their sources of inspiration and by some of the striking results generated in the visual realm. Technology on the right hand side of the brain does surprise. Most of us don’t normally think of computers as artists’ tools – but then we don’t think of the painter’s canvas, the writer’s pen, the musician’s instrument as “technology” either, although that’s what they are. On the other hand, new technology that embraces and promotes creativity is far from unusual in the context of i3 work. And while the three original i3 research programmes are now finished, that work continues. Phil Ellis and Tony Brooks highlight the impact of the CARESS project in terms of the new projects, new courses and expanded research lines it has given rise to. Articles in this issue share the premise that technology exists for the whole of human beings and the whole of their lives.This is a position also adopted in a recent document that forms part of the preparation for the EC’s next Framework Programme (FP6), and is the focus of another major article in this issue. Ambient Intelligence is a central concept in the IST component of FP6 — but can Ambient Intelligence be AmI, that is, humanistic and people-friendly?. Elsewhere in Europe a differently-organised community of researchers is consolidating research lines akin to those prominent in i3, with human-centredness as one of its chief directions. Patrick Purcell gives an update on some recent work at Media Lab Europe, which shows how “extreme” interfaces can get. Considering that i3 will come to an end in February 2003, it is hard to read Patrick’s report on Media Lab Europe without feeling a little wistful. As imaginative and inspirational as the idea of a Network of Excellence is, if it cannot be supported in the longer term then perhaps the traditional notion of one larger centre with potentially enduring funding has much to be recommended.

Links to the Web sites of the i3 CI projects can be found on: http://www.i3net.org/i3projects/links.html Information about the ESE projects can be found on: http://www.i3net.org/schools/

1

Feature

Designing for Homo Ludens Bill Gaver Royal College of Art [email protected]

‘All work and no play makes Jack a dull boy’ – popular saying The advent of graphical interfaces, more than 15 years ago, revolutionised the way we think about computers.The desktop metaphor had such a complete inner logic that it seemed to dismiss offhand the tedious call-and-response interfaces that had prevailed till then.The computer as concept expanded from a tool to a virtual environment, from a clumsy machine to a place for exploration and experimentation. Research on interaction also expanded, as people sought new perspectives from which to understand this newly-discovered territory – from cognitive, to perceptual, to ethnographic and anthropological. As our appreciation of computing’s potential grew, so did our appreciation of the aspects of humanity it mirrors. Now we are on the brink of another revolution, as computers invade our everyday lives.The point is not that computers are becoming ubiquitous or ambient or disappearing altogether. Nor is it that interaction will be tangible, or that the virtual will merge with the physical. These things may happen, but they’re symptoms — attempts to shortcut technologically the challenges we face. The real revolution is that computing is leaving the confines of task-oriented, rational, ‘left-brain’ work, and is set to join us in our homes, on the street, at parties, on lonely mountaintops – everywhere, in short, where ‘work’ is the stuff we want to get done so we can do what we really want to do.

magazine

The effects of this new revolution will, I believe, be as radical as the move to graphical interfaces, causing us to rethink computers, research, and even ourselves. However, we have been slow to appreciate the implications of technology’s incursion into our everyday lives. As computing has emerged from the office and laboratory, it seems to have brought along values of the workplace: concerns for clarity, efficiency and productivity; a preoccupation with finding solutions to problems. If, as ethnographers suggest, it takes a lot of work to achieve an ordinary life, then new technologies will help us take care of it. In this vision, internetenabled refrigerators will automatically update our shopping lists.We will use our microwave ovens to do our banking as well as to heat ready-made meals. Mobile devices will allow us to coordinate our schedules, download information, update records on the move. We will be surrounded by technology devoted to taking care of our everyday chores, giving us the leisure to pursue whatever activities we really value. But what if technologies helped us pursue those activities now, directly, rather than merely helping us get the chores done? What if computing helped us pursue our lives, not just our work?

‘We are here on Earth to fart around.’ – Kurt Vonnegut The idea of Homo Ludens – humans defined as playful creatures (Huizinga, 1950) – is an antidote to assumptions that technology should provide clear, efficient solutions to practical problems. From this perspective, we are characterised not just by our thinking or achievements, but by our playfulness: our curiosity, our love of diversion, our explorations, inventions and wonder. An aimless walk in the city centre, a moment of awe, a short-lived obsession, a joke – all are defining and valuable facets of our

2

humanity, as worthy of respect as planning, logic or study. Play is not just mindless entertainment, but an essential way of engaging with and learning about our world and ourselves – for adults as well as children. As we toy with things and ideas, as we chat and daydream, we find new perspectives and new ways to create, new ambitions, relationships, and ideals. Play goes well beyond entertainment: it’s a serious business.

What sorts of computational device might appeal to Homo Ludens? The Pillow, by Tony Dunne (Dunne, 1999; Gaver & Dunne, 1996; see Figure 1) might be an example. This is a clear inflatable pillow, enclosing a translucent plastic block with holes cut in it, exposing patterns of colour as they play over an LCD screen. As the light suffuses the object, it creates a much softer, more imprecise display than we usually associate with computers. But the Pillow isn’t just an aesthetic object; it is a strange form of radio in which lighting reflects bits of electromagnetic information from radio stations, passing taxis or nearby baby alarms. And more than a radio, it is a poser of socio-cultural questions, pointing out the degree to which our homes and even our bodies are permeated by wireless communications. It casts its viewers as meditative voyeurs, providing them with a gentle electronic experience while subtly eliciting unease about the communications that feed it. It is an object which invites a relationship, not as a pet, but perhaps as a sort of computational alien sharing one’s home (Dunne, 1999). Some of the designs developed in the i3 Presence Project, too, might speak to Homo Ludens (see Gaver & Hooker, 2001). At the RCA, Tony Dunne, Ben Hooker, Shona Kitchen, Brendan Walker and I explored ways in which technology might increase the presence of older people in three communities: a hilltop village in Tuscany, an affluent district of Oslo, and a huge housing estate in the Netherlands. Our designs took a wide and playful view of ‘presence’. In Italy, we proposed a ‘radioscape’ that would transmit sounds from the countryside to the village, encouraging the older people to enjoy their pastoral landscape in new ways (Figure 2). In Oslo, we suggested that older people might use a ‘digital boudoir’ to craft questions for their fellow citizens, to be displayed on trams or in cafés, or relayed to public phones (Figure 3). In the Netherlands, we proposed and built the Projected Realities system, which disseminated people’s attitudes from their private flats, through local neighbourhoods, to the roads and railways surrounding this notorious area. With visible elements including ‘sloganbenches’ (Figure 4a) and an ‘imagebank’ (Figure 4b), the Projected Realities system allowed passers-by to encounter the words and images of their older neighbours in a way that was not didactic or demanding, but quietly suggestive.

Figure 1:The Pillow suggests a kind of ambient voyeurism.

Figure 2: Peccioli’s Radioscape brought the countryside into a Tuscan village.

Figure 3:The Digital Boudoir suggested that older people could lead a political conversation distributed through Oslo.

3

Figure 4a:The Presence sloganbenches

magazine

Ludic appeal might also be found in the proposals that Heather Martin and I developed for the Alternatives project (Gaver & Martin, 1999). Funded by Hewlett Packard, the Alternatives project was part of the Appliance Design Studio, a collaborative investigation of information appliances. In investigating the field, Heather and I found ourselves uninspired by current examples, and developed a series of sketch proposals to expand the group’s thinking. For instance, the Dawn Chorus (Figure 5a) was a birdfeeder that would use operant conditioning principles to teach local songbirds new tunes. The (De)tour Guide would be an audio-only device using GPS and an electronic compass to lead people through the city – and to support them in getting lost for a predetermined interval. The Intimate View camera (Figure 5b), later developed as a prototype, linked separated lovers by allowing them to capture and transmit small, highly magnified pictures to encourage moments of intensely shared focus. The Dream Communicator allowed distant lovers to use sounds or speech to influence one another’s dreams. Finally, the Telegotchi was an electronic pet with no buttons that relied on psionic communication for happiness (Figure 5c), and the Prayer Device (Figure 5d) would be found on streets, like a new sort of telephone booth, waiting to transmit one’s voice to the sky. The appeal of many of these proposals, in particular, was that they didn’t demand belief so much as a suspension of disbelief.They encouraged an attitude of speculation that in itself might be enjoyable. The examples described here may be pleasurable to experience, but it should be clear that they go beyond mere entertainment. Each raises issues and

4

Figure 4b:The Presence imagebank

asks questions, ranging from the effects of pervasive electromagnetic communication, to possibilities for inter-generational communication, to the ethics of taming nature, the value of getting lost, or the status of psychic or spiritual experiences. They raise these issues, but don’t provide answers. Instead, they offer ways for people to experience life from new perspectives, thereby testing hypotheses about who we might be or what we might care about.They hint at possibilities for technologies that we could use in our everyday life, not to accomplish well-defined tasks but to expand in undefined directions. Openended and personal, they encourage us to play – seriously – with experiences, ideas and other people.

‘…work is play for mortal stakes…’ - Robert Frost What does designing for Homo Ludens imply for our methodologies? How can we invent and develop systems that legitimise wonder, even encourage it? How do we encourage people to meander, rather than to accomplish tasks with speed? First, scientific approaches to design need to be complemented by more subjective, idiosyncratic ones. It is difficult to conceive of a task analysis for goofing around, or to think of exploration as a problem to be solved, or to determine usability requirements for systems meant to spark new perceptions. Instead, designers need to use their personal experiences as sounding boards for the systems they create. Balancing this, they need to engage in, and often lead, a conversation with the people for whom they are developing, lest their

Figure 5:The Alternatives proposals suggested new information appliances. Left to right, top to bottom: Dawn Chorus, Intimate View,Telegotchi, Prayer Device.

designs become purely self-indulgent. Traditional requirements capture or ethnographic methods may be useful in this, but more ambiguous, openended forms of engagement can also produce inspiring results. For instance, the Cultural Probes developed for Presence (Gaver et al., 1999; see Figure 6) used provocative questions and tasks to elicit informative materials from volunteers. The returns were never definitive, but they were evocative, allowing us to create semi-factual narratives about the communities for whom we were designing, and to develop design ideas that furthered these stories. Second, designing for Homo Ludens means allowing room for people to appropriate technologies. Playing involves pursuing one’s inner narratives in safe situations, through perceptual projection or, ideally, action. If computational devices channel people’s activities and perceptions too closely, then people have to live out somebody else’s story, not their own (cf. Wejchert, 2001). This might be an interesting possibility – as Dunne (1999) suggests, people might approach computational devices the

Figure 6: Cultural probes are provocative materials eliciting inspiring responses.

5

way they do cinema, borrowing the identities and values they suggest for a short period of time – but in general we should give people the ability to own technology, to bring it into their own complex life stories. I know two primary tactics for doing this.The first is to create ‘suggestive media’ – suggestive in that they are designed to encourage or impel ludic activity, and media in that they are tools through which people experience, create, or communicate freely. The second is to employ ambiguity at all phases of design. Contrary to traditional thinking about interaction, ambiguity is an invaluable tool because it allows people to find their own meaning in uncertain situations. Used in design processes, concepts and products, ambiguity gives space for people to intermesh their own stories with those hinted at by technologies.

magazine

Third, and most important, pleasure comes before performance, and engagement before clarity. Designing for Homo Ludens requires a new focus that seeks intrigue and delight at all levels of design, from the aesthetics of form and interaction, to functionality, to conceptual implications at psychological, social and cultural levels. Not only should technologies reinforce pleasures that people know, but they should suggest new ones. The designer’s role in this is not like that of a doctor, prescribing cures for people’s ills; nor is the designer a kind of servant, developing technologies that people know they want. Instead, designers should be provocateurs, seeking out new possibilities for play and crafting technologies that entice people to explore them (cf. Dunne 1999). In the end, designers themselves need to be Homo Ludens.They need to recognise that they are playful creatures, and that their work depends on their play. References Dunne, A. (1999). Hertzian tales: Electronic products, aesthetic experience and critical design. London: RCA:CRD Research Publications. Dunne, A. and Gaver, W. W. (1997). The Pillow: Artist-designers in the digital age. Conference Companion for CHI’97. Gaver, W.W.. Dunne, A. and Pacenti, E.. (1999) Cultural Probes. interactions magazine. vi(1), pp. 21 - 29. Gaver, W. and Hooker, B. (2001). The Presence Project. London: RCA:CRD Research Publications. Gaver, W. and Martin, H. (2000). Alternatives: Exploring information appliances through conceptual design proposals. Proceedings of CHI’00 (Den Haag). New York, ACM Press. Huizinga, J. (1950). Homo Ludens: A study of the play-element in culture. Boston:The Beacon Press. Wejchert, J. (2001).The Dreaming. Informatik/Informatique 5/2001.

6

Acknowledgments

I am grateful to Jake Beaver, Anne Schlottmann, and Jakub Wejchert for their comments and suggestions.

Bill Gaver has a background in psychology and politics, and has pursued research on topics ranging from everyday listening to technology-mediated sociality. Since joining the Computer Related Design department at the Royal College of Art (www.crd.rca.ac.uk), he has been using design as a means of research, exploring new cultural and emotional territories for digital devices. From 1997-99, Gaver and Tony Dunne led the RCA team on the i3 Presence project, working with partners from four countries to explore innovative interaction techniques to increase the presence of older people in their local communities. He later headed CRD’s involvement in the Appliance Design Studio, a virtual organisation spanning Hewlett Packard, IDEO, and Appliance Design Ltd. Currently he is a principle investigator in the Equator IRC, collaborating with seven U.K. universities to develop new ways to merge the virtual and the physical. His group is focusing on ludic technologies for home.

Feature

Putting the art in artificial What can computing bring to the study of creativity? Creativity is a trait of humankind which is treasured at least as much as, if not more than, any other. Without exception, human societies value individuals and groups who create, whether the intent be art or craft (or both), ritual or entertainment (or both), serious in purpose or just for fun.

Geraint Wiggins City University, London [email protected]

It is often suggested that creativity is also a uniquely human trait – that is, that other animals do not exhibit it.This claim is perhaps yet to be verified, and it is certainly rather inscrutable: it is notoriously difficult to define which acts are “creative” and which are not, let alone to determine whether a process encoded as nerve impulses in the brain of a given organism can be called creative or not. Perhaps this inscrutability is one of the reasons why adherents of the study of Artificial Intelligence (AI) and Cognitive Science have been slow to take up the study of creativity and/or creative behaviour. In spite of brave (and radical) protestations by Margaret Boden in the mid-70s, it is only since about 1995 that the study of creativity in itself — as opposed to that of creativity exhibited as part of a more concrete task — has taken off. Computers and creativity What is the general motivation for using a computer in the study of creativity? Broadly speaking, there is a spectrum of activities, or interests, in AI, ranging from (at what I shall call the “engineering” end) the attempt to make computers more like people so that, for example, people can interact with them better, to (at what I shall call the “psychological” end) the attempt to understand the human mind better by so-called cognitive modelling. Between these two extremes lies a continuum with various degrees of emphasis between the two viewpoints1 . All these activities, whatever their position on the spectrum, are at least partially aimed at better understanding the human mind.Therefore, the study of creativity and/or creative behaviour must surely have a place amongst them. And perhaps we may also suppose that, as a phenomenon, creativity may be amenable to some of the same methods of study to which other human mental traits have yielded in the past. If so, one important aspect of such a scientific approach is the idea of formalisation. Formalisation is the process of understanding and describing a problem in such a way as to remove all ambiguity from our description. One consequence of formalising a problem is that we are subsequently better placed to write a computer program to solve it, because computers are much better able to deal correctly with problems stated in formal terms than with ones stated informally.

Moreover, such a formalisation has spin-offs outside the computational world. The aim of better understanding a natural occurrence such as human creativity is no different from the aim of abstract philosophy in the same area. So the computational study of creativity runs alongside a philosophical understanding of the same phenomenon. In terms of research on creativity, the above spectrum ranges from the study of creative computer systems on the one hand (the “engineering” or “doing” end)2 to the computational study of creativity itself on the other (the “psychological” end). In what follows I will discuss two well-known examples at different ends of the spectrum. AARON The first example, at the “doing” end of the spectrum, is that of Harold Cohen whose artificial artist, AARON, is perhaps the most famous of all creative computer systems3. AARON is a vast computer program, developed by Cohen (himself an artist of significant standing) more than 30 years ago, which is capable of driving a robotic painting arm to produce paintings4. AARON produces original paintings. It clearly has its own style and does not repeat itself. Its paintings are valued enough, at least by some sections of human society, to be placed alongside the output of human artists in galleries – and not just as a freak show.

7

However, from the research point of view,AARON’s value is limited.The reason for this is that its creator chooses not to explain to the research community how it works – he merely explains what it does. This, of course, is his prerogative, but it does make it hard to answer the most interesting question one could ask about AARON: Is it really being creative? Cohen himself has changed his opinion on this: for many years he denied that AARON was creative at all, but recently he has capitulated as far as acknowledging that AARON is “creative with a small ‘c’”. As I understand it, Cohen’s unwillingness to attribute creativity to his amazing program arises at least in part from his understanding of its limitations (which he necessarily understands better than anyone else). Specifically, AARON has a limited repertoire of picture elements: all its paintings are of rather angular people placed around large plants in pots, in otherwise empty rooms. AARON invents new scenes from these building blocks, but cannot autonomously add new forms to its visual vocabulary. But does this limitation mean that AARON is not being creative? Zara

magazine

Cohen is very proud of his little daughter Zara, and has written about her development as an artist. There is little doubt, when one sees her drawings, that she has talent and potential and that she is very creative — like many, many children. But this raises an important question: when we compare AARON’s output with that of, say, a fouryear-old, it is immeasurably more controlled, consistent and (significantly) structured; and since AARON’s paintings are all newly generated they are also by definition original.These are all attributes that are generally valued in artists’ work.Yet we still have a problem with accepting that AARON is creative because it is a program and not a person. This is so even though AARON’s output is apparently more valued as art than the drawings of a four-year-old, who all may agree is definitely creative. We seem to be contradicting ourselves. Creativity and value One way to begin to resolve this paradox is to pinpoint its source, and Margaret Boden’s work on creativity – my second example, at the “psychological” end of the spectrum — helps us do

8

so. Boden is a psychologist and philosopher with a gift for explaining things in an accessible way, as her book, The Creative Mind: Myths and Mechanisms5, testifies. Over an extended period, Boden has attempted to clarify the discussion of creativity in philosophy (and hence in AI) by codifying different categories of creativity and trying to characterise the behaviours associated with them. One of Boden’s most important arguments is that behaviour can be creative independently of how the result of that behaviour is valued. This distinction between creation of an artefact and the value we place on that artefact can perhaps help us with our paradox. In particular, it can demystify creation itself: if evaluation is somehow independent of the creative act, we are newly allowed to be creative but to create things which are not valued highly. In other words, it is not only the great composer recorded by history who is creative; the teenager in his bedroom with his electric guitar is creative too.The success of creative behaviour can perhaps be measured in terms of its product; but even if it is not rated as successful, it is still creative. How exactly does this help? It allows us to say that little Zara is creative without necessarily (yet) valueing her output as significant, which removes half of the paradox. But it still leaves the converse question unanswered: how can it make sense to evaluate AARON’s output highly (as we do) under criteria which are intended to rate the creativity of human artists, and still claim that AARON is not being creative itself? Two ways of designing a bungalow Boden gives us another tool with which to chip away at this problem. She attempts to characterise different kinds of creativity and to compare them. The most important distinction is between what Boden calls exploratory creativity on the one hand, and transformational creativity on the other. To understand these ideas, we need to share Boden’s view of the creative universe. Boden proposes that the act of creation consists of picking concepts (that is to say, created artefacts) from a range (called a conceptual space) of possibilities.This range contains all possible concepts related to whatever context we are creating in. The creator doesn’t have access to all of them at the same time because, in any interesting context, there are just too

Figure 1: Part of the conceptual space of bungalows, showing the boundary, defined by the rules that say what a bungalow is. A completed bungalow (A, B and C are complete in this simple diagram) may be reached from the empty concept by more than one chain of reasoning, as shown by the arrows (A is an example). However, the two-storey house, H, is not reachable from the empty concept, because the boundary (which represents the limit of “bungalowness”) is in the way.

many to think of all at once. Some of the possibilities in the range may be valued as being very important and some may have no value at all. But regardless of this, the process of creation is akin to searching through the conceptual space. The boundaries of the conceptual space are defined by rules that say what is an artefact in the context to which we want to apply creativity. Again, quality is not an issue: an artefact is either relevant, or not. An example will help make this clearer. Consider the conceptual space of bungalows. It contains the empty concept (which we can think of as “a bungalow that hasn’t been thought of yet”), and it also contains partial concepts, like having five rooms, or having a red roof and four walls and a front door6. Finally, the conceptual space contains

complete concepts – bungalow ideas that are precise enough to use as builders’ plans. The process of creating a new bungalow can be seen as moving around the conceptual space, stepping from (partial) concept to (partial) concept until a complete concept is reached. We need to work like this because the space is so large that we can’t just go directly to the bungalow we want – we simply don’t know where it is. But we know when we have found it, because there won’t be any parts missing from its specification. This search for a finished concept is illustrated in Figure 1; bungalows A, B and C are complete concepts. There are lots and lots of bungalows we could create like this (most of them are not shown in Figure 1!). Some are good and some are bad. Some

9

Figure 2:The transformation from an example defining rules set for bungalows to that for more broadly defined buildings.The dotted arrows show where the change is.This actual change is much smaller than the changes indicated by the arrows in Figure 1, but its effects are much more profound.

magazine

might have windows in silly places. Some might be prettier than others. But no matter what their properties, thinking them up is still creative – it is the bad product that is flawed, not the creativity. In Boden’s terms, the selection of the nicest bungalow is evaluation; but creating any bungalow, no matter how ugly or impractical is exploratory creativity. Moving up-market So what is transformational creativity? In my bungalow example above, one of the rules defining the conceptual space is that there is exactly one storey in the building. An example of transformational creativity would be to replace that rule with one requiring at least one storey. This would mean that our entire conceptual space was changed: we could have houses with as many floors as we liked. This is clearly a bigger deal than just designing a new bungalow, as it allows a drastically more efficient use of ground space. The changing rules are shown in Figure 2; as a result of the change, house H in Figure 1 (and many other multi-storey buildings) is now included in the new conceptual space.

10

In abstract terms we can summarise the difference between the two kinds of creativity as follows. Exploratory creativity merely locates new concepts in the range of possibilities; in principle, we already knew they were there, but we hadn’t noticed them yet. Transformational creativity, on the other hand, involves changing the rules that define the range of available possibilities, making the range larger or smaller, adding new and removing old possibilities. So completely new ideas, which could not have been thought of before, may be possible. Boden suggests that transformational creativity is inherently more important than exploratory creativity, because it is transformational creators that “change the world” by making it possible to discover new ideas. However, and perhaps confusingly, this is still different from the evaluation question: the product of transformational creativity can still be totally valueless – the point is that transformational creativity is somehow more fundamental than exploratory creativity.

Creativity with or without a capital Now, though, we can begin to get a handle on what Cohen means when he describes AARON’s creativity as having a “small ‘c’”.What AARON does is more or less exactly Boden’s exploratory creativity: it has very strict constraints on what it can do, and a limited repertoire of “construction blocks” to build things from. AARON’s empty concept is a blank canvas; a complete concept is a collection of AARON’s visual building blocks (as defined by Cohen), put together on that canvas, in a way that conforms to AARON’s painting rules (also defined by Cohen). But AARON is not capable of transformational creativity – it cannot create, say, a picture involving an elephant, because it does not have any rules defining what an elephant looks like; it cannot even create something completely abstract out of random numbers, because it does not know how. Where does all this lead us? It gives us some very rudimentary mathematical tools for describing and understanding creativity as a computational process. But more broadly, it has already led to computer programs capable of near-human-quality artistic and scientific output. It has led to computer programs which very effectively assist humans in producing their own artistic and scientific output. In the longer term, it may lead to computer systems which are capable of independent creative output, valued as much as that of human creators. And ultimately, it may help us understand the creative behaviour we are so proud of in ourselves.

Footnotes 1 It is worth adding that it is fairly unusual in the current AI world to find a researcher who claims that his or her work is actually creating intelligence – rather, we are interested in simulating it in various degrees of detail, for reasons that can be placed somewhere on the spectrum above. 2 Amilcar Cardoso reports on this elsewhere in this issue, see pp 20 – 26. 3 See also Linda Candy’s article on COSTART, pp 12 – 16 of this issue. 4 An on-screen version of AARON has recently been made generally available and can be downloaded, on a free trial basis, from Kurzweil CyberArt Inc. Unfortunately, the licence precludes an example here. 5 Published by Abacus, 1990, ISBN 0 349 10469 7. 6 These partial concepts are not the same as partially built buildings – we can have the concept of just “a bungalow with a slate roof ”, but we’d run into trouble if we tried to build it literally, because there would be no walls to support that roof.

Dr. Geraint A. Wiggins is a senior lecturer in the Department of Computing at City University, London, where he is currently leading the institution of a new Centre for Computational Creativity. He has been engaged in the study of computational creativity since 1989, through an interest, initially, in computer music and, later, in creativity in general. He is chair of AISB (the Society for the Study of Artificial Intelligence and the Simulation of Behaviour), the UK learned society for AI and Cognitive Science. For more on computational creativity:

Recommended Reading Boden, M. The Creative Mind: Myths and Mechanisms. Abacus, 1990.

http://www.soi.city.ac.uk/~geraint/

Koestler, A. The Act of Creation. Hutchinson, 1976.

11

Feature

The art of collaboration A guide to co-creativity in Art + Technology Linda Candy Loughborough University [email protected]

When artists and technologists collaborate, who is in control? Should technologists support the artists or should they be equal partners? In artist-in-residency studies, these have emerged as vital questions. Linda Candy explores them and describes what has emerged from relevant work at the Creativity and Cognition Design Studios. Art and digital technology have only relatively recently come of age. Up till the late 1960s, a few pioneers were working largely in isolation and with little recognition that what they were trying to achieve bore any relationship to art practice. Individuals and small groups worked at the boundaries between art, science and technology, very much against the prevailing cultural wisdom of the traditional arts world: their efforts were more likely to be viewed as crude manifestations of science fiction than as art.

magazine

In 1968 a landmark exhibition, ‘Cybernetic Serendipity’, took place in London and became the first public demonstration of the state-of-the-art in art and technology. It was around the same time that AARON,1 the pioneering automatic-art-creating computer program written by the painter Harold Cohen, was born. People hailed the new “infant”, who could draw only in black and white at that time, as a signal that autonomous machine creativity was now truly with us and was set to challenge our notions about the uniqueness of human creativity. Since then AARON has grown up: he is over 30 years old now, and has graduated to colour and complex figurative compositions. Interestingly, his human creator does not claim that what the program does is “creative”:

“I used the term “new, original images,” not “creative.” I use the word “creative,” on those rare occasions when I use it at all, to refer to the ability of the individual – human right now, program potentially – to move forward, to develop, to introduce new material. These imprecise terms give a flavour, not a definition. To put it more precisely, I believe the word properly attaches to continuous change, not to single events. There is no question that AARON has moved forward

12

and developed over its 30-year existence, but the agency of change and development has been me, not AARON.” Harold Cohen,A Self-Defining Game for One Player So, whilst AARON is now a better painter, more importantly, Cohen is a different artist. Today, the world of art and technology has changed remarkably since those early beginnings and there is much more diversity in the new generation of digital creations. What was once conceived of as “computer art”, where the control of the creative process and the making of artefacts belonged to the machine, has given way to new ways of working both with digital technology and with other people, where the collaborators may be human, computer, or both. We call this co-creativity. Practice-based research in Art + Technology The idea that interaction with digital technology plays a significant role in the creative process has driven the work we have been doing at Loughborough University for some time. We want to understand how this process works and what it tells us about the nature of creativity in the context of art + digital technology. However, opportunities to investigate such events are limited, and we want to avoid laboratory-style experiments that might strip the evidence of the richness that real events provide. Our approach was to establish the Creativity and Cognition Research Studios (C&CRS) for exploratory Art + Technology projects, and to make artist residencies the vehicles for research into the creative process.

C&CRS provides an environment where artists and technologists can work as collaborators.The Studios were established in 1999 as a joint venture between the School of Art and Design and the Department of Computer Science. New initiatives in creative digital art provide a mechanism for informing our general understanding of the creative processes and the role of the technologies involved. In the COSTART project, funded by the Engineering and Physical Sciences Research Council (EPSRC), expertise in both Human Computer Interaction and Digital Art combined to support residency projects. The starting point was to ascertain what the artists’ needs and expectations were, not only in terms of the technology required but also in terms of the skills and knowledge of other experts.

events as they happened in field diaries and discussing the work during the artist’s time in residence as well as during the post-residency fellowship period. We are now just half-way through the project and have learnt much already that has made us reexamine some of our initial assumptions. The ideas that led the practice-based research approach, along with first hand accounts by the artists themselves, are set out in a new book, Explorations in Art and Technology (Candy and Edmonds 2002). To give a flavour of the projects, here are two examples by artists Mike Quantrill and Anthony Padgett. Observations

“The main support I observe artists needing is that of Artists were identified from the respondents to a survey and were invited to a workshop to discuss requirements with prospective support staff and researchers. The idea was to conduct a series of artist residencies for which expenses were allocated in the manner typical of artist bursaries. In addition, each artist was made a visiting fellow, to enable them to continue their work for one year after the residency itself was completed. It was particularly important that artists were prepared to participate actively in the research. This involved recording

people support. It is not enough to have systems that artists can use, they need real contact with people who understand the technologies and can effectively communicate with the artists.These people would be more than technicians. For the best results they would need to be sympathetic to the artists’ concerns and not just interested in solving technical problems”. Michael Quantrill: Integrating Computers as Explorers in Art Practice

Project 1: Drawing with a SoftBoard The Soft-Board is a whiteboard that connects to a computer and is designed to support business meetings. The whiteboard is similar to a conventional whiteboard, except for the fact that it has a laser matrix across its surface area. The matrix enables pen and position data to be transmitted to the computer. The Soft-Board does not use an input device (such as a graphics tablet or mouse), so the artist has complete freedom of movement.This freedom allows the work to evolve in a way that overcomes some of the constraints usually associated with electronic media.A number of artists have deployed the Soft-Board for different artistic purposes.

Figure 1: Mike Quantrill working with the ‘SoftBoard’ (COSTART project)

13

Project 2: Interaction and the Sensor System The Sensor System consists of an array of infrared position sensors which are located in a physical space and connected to a computer. A display is projected onto a screen on one side of the grid and a sound system is connected to the computer. People moving in the space do not have to be conscious of the computer: they can move about freely and as they do so, the computer transforms their movement into sound and visual representations that are projected onto the large screen.

Figure 2: Anthony Padgett and the Sensor System (COSTART project)

Learning new skills and techniques is an important facilitator of creative practice. If the artist does not have the skills, the role of a collaborator is essential. Some artists may want to take full control of the technology because it is pivotal to the way they work, whilst for others a temporary need can be met by a technology expert.

magazine

However, technologists with little knowledge of art practice do not easily make good collaborators. Artists need collaborators who understand their need to exercise control for themselves. Working through the eyes and hands of the person who provides technical expertise is not right for the core creative activities, although it may be acceptable for the more mundane ones. Support in itself is not enough The first major lesson that came out of the initial residencies was to do with the concept of “support” itself. In response to artists’ demand for technological facilities and expertise, the preparation for the residencies concentrated on two things, namely:

For the technology, we had an established base of “high-end” computing equipment, network facilities and a repertoire of office and drawing software, as well as specialised packages for 3-D modelling and a position-sensing system. Where a specific piece of technology was needed, that was acquired for the purpose of the residency project. However, whilst some of the artists had well-developed skills in the use of some technology (because the projects were set up with a view to exploring new digital forms,) we did anticipate the need for help from experts in more advanced technologies. To achieve this, we established a network of willing experts inside the university. We envisioned artists driving the projects and technical people supporting the process in response to the artists’ requirements. This did happen, and the programmers in particular found their skills in 24-hour demand for the duration of the residencies. Even so the support provided was, in fact, never really enough although there was no doubt that the artists appreciated the time and commitment that was given. Extending support from assistance to partnership

• the technology itself, that is, the required software packages and hardware devices needed to carry out the artist’s project; and • people with the technical knowledge to enable the use of that technology.

14

From the artists’ point of view, having access to support for specific activities, such as programming or digital video editing, was only part of the story. One of the interesting things we observed was how much further the artists themselves wanted to

extend the support of their “assistants”. A significant number were looking for more than technical knowhow and seeking a partner for artistic exploration. For that to work, the assistants needed to engage actively in the creative process and resist imposing a standard technical solution. Likewise, the artists needed to be open about their intentions and be prepared to reveal tentative ideas that would normally remain hidden until they reached a more mature state. For a large part of the time spent in the Studios, the collaborators discussed their work in pairs whilst seeking out more private spaces for individual tasks such as programming or shooting video clips. A pattern of regular, planned contact rather than intermittent, chance meetings seemed to indicate a more productive collaboration style. Success factors for co-creativity The COSTART residencies identified several issues about the nature of art-technology co-creativity. Partnerships tend to be most successful when they serve convergent interests but produce distinct artistic outcomes, so that the partners achieve mutual benefit but also retain ownership of their individual achievements. Being able to enjoy such mutual benefit requires the relinquishing of individual control of the creative process, and having different but complementary roles appears to be best suited to achieving that end. Having respect for differences in methods is also important for a successful partnership. The trick is for the people concerned to identify in what way differences in approach can be mutually beneficial, and when they do not really matter in the larger scale of events. If a partnership is perceived as art-led by both parties, this leads to a better relationship. The technology-led situation, on the other hand, may have the effect of placing the non-technologist at a disadvantage, both in terms of control of the creative process and in terms of the eventual outcomes – the art forms. If the implications of adopting a particular technology solution are not fully understood by the artist, it may not be possible to steer the direction of the work to suit, resulting in a loss of artistic control.

For the technologist, the disadvantage of an art-led situation lies in a lack of ownership of the project. This may occur even if the technologist is providing critical input to the process through such contributions as (say) programming design. Where the relationship is of the assistant type, it is more productive if the artist explicitly acknowledges the value of the technologist’s contribution and actively tries to learn from it. Sharing knowledge is an important facilitator of creative collaboration. If both partners can exchange knowledge resources in order to move the work forward and resolve both technical and artistic difficulties, this results in an effective working relationship, particularly if the parties have complementary skills. Moreover, a partnership that aims to be self-sufficient must also know its limits and be willing to carry out the necessary research when the knowledge proves to be insufficient. Indeed, the quest for self-sufficiency in technical know-how, through research, can be a stimulus for creative thought. Being able to learn through knowledge-sharing is beneficial, particularly where direct contact with a new way of thinking stimulates the generation of options. In one such case, as the process of programming became clearer, the artist was able to understand more fully the basic logic. This enabled her to consider more carefully what her options were and how the aesthetics of the piece could operate. For successful partnerships, a longer-term relationship in which trust and confidence can be built up has real advantages. But there are many potential communication barriers. Openness, flexibility and a willingness to engage fully support communication; a lack of flexibility may suggest unspoken differences about the way the project is developing. Communication difficulties sometimes reflect a different way of thinking about the problem in hand. Developing a common language (particularly when discussing technical issues) that both parties can understand and work with is essential. In the case of an “assistant” style of collaboration, there tended to be more difficulties in finding a shared vocabulary.

15

Conclusion: successful collaboration can be learnt It follows from all this that learning how to collaborate successfully is very important to cocreativity in art and technology and cannot be assumed to come naturally to everyone. We can facilitate successful collaboration by making the conditions as suitable as possible, but we need to be aware of the critical human issues at play. Media centres — and any organisations wishing to promote collaborative creativity – need to pay more attention to developing learning strategies for successful collaboration. To achieve this, understanding how good partnerships work is necessary, as is providing the means for sustainable relationships. A number of artists have continued their association with the C&CRS and the second phase of the COSTART project is now underway.To be successful over time, creative partnerships need appropriate organisational support. An environment that supports co-creativity involves much more than the choice of technologies and technical skills, vital though that remains. Fostering co-creativity through sustainable partnerships is a key requirement for successful art + technology collaboration.

Key requirements for sound and productive partnerships: • devising a shared language and foster its development; • developing a common understanding of the project’s vision and its specific artistic intentions; • engaging in intensive discussions and “what if?” sessions; and • making time to establish the relationship and to recover from misunderstandings. Artists need: • a network of resources for a broad range of requirements and approaches; • access to high-end facilities and tailorable digital systems; • access to appropriate human expertise that is communicated well; and • an ability to reflect and learn from technologists. Technologists need: • good communication skills;

Footnotes

magazine

1 For more on Aaron, see also Geraint Wiggins’ article on pp 7 – 11 of this issue.

Dr. Linda Candy is Senior Research Fellow in the Department of Computer Science at Loughborough University. She has considerable experience in research into creative uses of computers and is currently principal researcher for the EPSRC project ‘Studies of Computer Support for Creative Work: Artists and Technologists in Collaboration’. Her main research areas include creativity research, interaction design and methods for usability evaluation, and she has published widely on these topics. She is co-chair of the International Symposium on Creativity and Cognition, sponsored by the ACM Special Interest Group on Computer Human Interaction, and of the 3rd International Conference on Strategic Knowledge and Concept Formation. She has been invited to present her work in Europe, Japan, Australia and the USA. For more information about Creativity and Cognition Research Studios, Loughborough, UK and the COSTART project: www.creativityandcognition.com.

16

• an ability to listen and learn from listening; and • an ability to suppress the urge to promote a course of action that is technically feasible but not artistically valid.

References Cohen, H: A Self-Defining Game for One Player. In: Leonardo Journal: Special Section on Creativity and Cognition, Volume 35, No 1, February 2002. Candy, L and Edmonds, E: Intersection and Correspondence: Explorations in Art and Technology. Springer, 2002. Quantrill, M: Integrating Computers as Explorers in Art Practice. In: , L.Candy and E.A.Edmonds: Intersection and Correspondence: Explorations in Art and Technology. Springer, 2002.

i3 legacy

Creating Aesthetically Resonant Environments CARESS The CARESS project began in 1998 and ran for 18 months, with a three-month extension. It was designed to enable young children — both children with special educational needs and children in mainstream primary schools — to learn and develop physical and cognitive skills by interacting with a responsive sound environment. In this second instalment of our series on the legacy of i3, Phil Ellis and Tony Brooks look at the impact of CARESS beyond the lifetime of the project.

CARESS

CARESS legacy

The technology that provided the basis for the CARESS project was the Soundbeam. The Soundbeam converts physical movement into MIDI (Musical Instrument Digital Interface) code: sound parameters can be adapted to movement parameters, resulting in a coupling of gesture and sound. For instance, a slow movement in the beam may trigger a much quieter sound than a fast movement, or an upwards head movement could trigger an ascending sound scale, depending on how the user chooses to program the Soundbeam.

Three projects based at the University of Sunderland are taking the ideas and techniques of Sound Therapy further.

Two sensors were developed as part of CARESS: a muscle sensor and an optical fibre wireless sensor. The project took an iterative approach to the design of these sensors, with pupils, support staff, teachers and educational researchers all contributing to the design process1. Both current sensors point to possible new developments, for instance in enabling much greater control and freedom of movement. In the special needs domain, the resulting nonintrusive Sound Therapy has improved children’s mobility and control over their environment (also outside the Sound Therapy sessions) and led to an “awakening”, observed in facial, vocal and body expressions — often of joy. Sound Therapy has also been successfully adapted and applied to aspects of the mainstream primary school curriculum, and the project has developed curriculum materials that will be circulated widely.

Phil Ellis University of Sunderland [email protected]

The first, entitled The Music of Sound, involves nonspecialist teachers and carers at two special schools exploiting the techniques developed in CARESS in their work with special needs children aged between six and 16. The project was designed to enable a number of teachers/support workers to receive training in Sound Therapy, so that the programme could continue and develop without the need for specialist support from outside the school. This was successful, and three members of staff at the school are now working independently with a range of children. This will enable the school to become a focus for other interested schools/teachers in the reqion. This project shows that one of the original aims of work in this area — namely to develop an approach that did not require intense specialist training, either in teaching and learning methodologies or in the use of complex or unwieldy technology — is now becoming a reality: the current technology and learning techniques are readily accessed, even by selfconfessed technophobes!

Although CARESS is now finished its ideas and results are very much alive. Three new research projects, all based at the University of Sunderland, illustrate this. Some of the CARESS work also lives on in another i3 project, CARE HERE (p19).

17

Figure 2:Thanks to Marc Blasband, coordinator of the i3 Partnership Programme, CARESS and Sound Therapy were able to be present at major recent exhibitions in Paris and Amsterdam (see i3mag 10, pp 24 – 25). This photo shows a group of people at the Amsterdam exhibition enjoying spontaneous improvisation. Figure 1: Exploring and expressing through physical movement motivated by sound. A second project has further developed the techniques of Sound Therapy for the elderly, supported with funding from the Linbury Trust. It follows up an earlier pilot study on technology for the elderly in long-term care. Research on the project has been located at three sites, all long-term care homes for the elderly mentally infirm. Over 20 people, with conditions ranging from mild/severe stroke and depression to dementia and extreme anxiety, have been involved.

magazine

The project has further explored the use of Soundbeam and interactive communication through non-verbal means and vibro-acoustic techniques. It is now nearing completion, with very positive outcomes. The techniques and technology explored and developed in CARESS are proving very effective in this challenging and very moving domain, and the next year of the project will potentially have a very real impact on the quality of life of some of the people involved. Additional technology and techniques in the domain of vibro-acoustic equipment (Soundbox and Soundchair) are being explored and developed in both these projects. Finally, a new course in dance and technology has been developed at the University of Sunderland, which is exploring possibilities for combining movement with the expressive creation of sound. Some potential for this was revealed during the CARESS project, but the CARESS foundation research on movement with primary school children is now being significantly extended; and, as in CARESS, it is students who are designing and discovering new ways forward as dancers by using the technology in different locations and contexts. This is giving rise to new possibilities, including ideas for new types of sensor.

18

A new research proposal is currently being prepared which would capitalise further on these possibilities, together with the exploitation of emerging wireless technology and new ways of giving more interactive and fluid access to sound.

In short Although CARESS was a short project we managed to fulfil its aims. And in doing so we fortunately generated more questions than answers — as often happens in research.These new questions offer many avenues for further exploration and development, research and dissemination. Being a member of the i3 community provides additional stimulation and the possibility of forging new links, in order to design, and maybe implement, new projects. Footnotes 1 Where children are involved from the start of a project and listened to, and their ideas and suggestions acted upon, the results are likely to be more useful and pertinent.This point was reinforced by the immensely useful workshop provided for CARESS by the i3 ESE working group Kidslab, held at the University of Limerick in December 1999 (for a report, see i3mag 7, pp 24 – 25). This feature of the i3 programme is very helpful and useful, and powerfully reinforces the i3 emphasis on building a community.

Phil Ellis was the co-ordinator of the CARESS project and supervisor of its activities at Sunderland. He is professor in Music in the School of Arts, Design and Media at the University of Sunderland. He developed the Sound Therapy methodology through his work with special needs children, and he serves as co-chair of the International Society for Music Education (ISME) Commission on Special Needs, Music Therapy and Music Medicine. He is introducing and evaluating the new technology at the Special Needs school in Stratford-upon-Avon and supervising the trials of the Soundbeam and the new technology at the mainstream school in Henley-in-Arden. Together with Lisa Percy, he is also responsible for developing curriculum materials resulting from the CARESS research. For more on CARESS: www.bris.ac.uk/caress

CARE HERE Another aspect of the i3 legacy comes to life in the i3 project CARE HERE, funded by the EU IST programme, which takes further the control through movement of both audible and visual material, often as a therapeutic aid. It is based on the Soundscapes concept and methodology, developed by Tony Brooks in the mid-80s (see feature in i3mag10, pp 2 – 6).

Tony Brooks Soundscapes [email protected]

In 2001 the merger of Soundscapes with the core of the CARESS team resulted in a six-month i3 “Future Probe” research project which further investigated the Soundscapes theories on using synchronised visual multimedia feedback (in addition to the “just sound” feedback from the CARESS project).The success of this probe led to CARE HERE. Tony was recently invited to present CARE HERE in New Zealand and Australia, and the pictures below are from this tour.They illustrate how it is possible to create Aesthetic Resonant Environments on a large scale and for a large cross-section of the public, including disabled and elderly people and those in rehabilitation.

Lucy, wheelchair-bound and a member of the leading New Zealand mixed-ability Dance company Touch Compass (www.touchcompass.org.nz) is moving within a Virtual Interactive Space© created by Tony’s 3D sensors and cameras, on the scene of the 1000-seat Dorothy Winstone Theatre in Auckland. She is manipulating sounds and "painting" through her movement.

An image from the Four Senses concert performances with Auckland Symphony Orchestra. Movement from the orchestra is captured and manipulated in real time with Eyesweb software (from Genoa University (DIST), one of the CARE HERE research partners; see www.infomus.dist.unige.it/). Behind the orchestra a live capture can be seen of the double bass musician’s hand (back screen top left) bowing the strings.This image is then blended into other images to suit the musical piece (in this case Rites of Spring).

Tony Brooks’ real-time improvisation with the sensors and cameras from the control box.The system created in New Zealand consisted of five computers, seven LCD projectors, five static cameras, three roving cameras/camera men on headphones, video mixers, a video feedback system and lots of analogue bits and pieces for fun.

19

Feature

Darwin and creationism reconciled When computers should surprise us Amílcar Cardoso University of Coimbra [email protected]

Could computational approaches to creativity facilitate the understanding of creativity? Is it worth investigating computational tools and environments that might help humans being creative? And is it feasible to build programs that we could classify as “creative” — or are computational approaches to creativity a waste of time and resources? It depends on the view you take of what creativity is. Does it find its origin in some kind of divine inspiration or innate talent (what Margaret Boden (1990) called the inspirational or romantic views of creativity)? Or is it a fundamental part of human intelligence, one of the most remarkable characteristics of the human mind? Like many others, the authors of this article take the latter view, and argue that, on such a view, it is inconceivable to research intelligence (natural or artificial) without studying creativity.

Carlos Bento University of Coimbra [email protected]

magazine Penousal Machado Politechnic Institute of Coimbra [email protected]

Francisco C. Pereira University of Coimbra [email protected]

Creativity is hard to measure, observe and interpret. Its study has been a challenge for many scientists and researchers, particularly for those from areas such as Cognitive Science and Psychology. In recent years, the subject has attracted a growing number of AI researchers who have been working towards abstract explanation theories and adequate computational models of creativity. This interest comes from the belief that computational creative systems are potentially effective in a wide range of artistic, technical and scientific domains where innovation is a key issue. Scientific discovery, theorem proving and technical design are just a few examples of application problems suitable for them. Moreover, the endeavour may contribute to the overall understanding of the mechanisms behind creativity. Process and product When studying creativity it is useful to consider two distinctive, complementary aspects: the creative process and the creative product. The creative process is central to creativity modelling, and several explanatory models have been proposed for the human creative process (see Brown (1989) for a survey). Models adopting an information processing approach (e.g. Wallas (1926)) are particularly useful for computational creativity. Roughly speaking, these models describe the process as a stepwise procedure of problem acquisition and knowledge assimilation; conscious or unconscious search for a solution; proposal of a solution; and verification of the proposed solution. But analysis of the creative product is also of prime importance, as creativity is most often recognised on the basis of its outcome: a symphony, an invention, or a theorem proof are commonly accepted as creative

20

products on the basis of what they are, rather than the specific process that produced them. Although we may not agree with a view of creativity that focuses solely on the product, it is hard to imagine an assessment that does not consider the product as one of its main components. Two main properties are accepted as characterising a creative product. One is novelty: for a product to be recognised as creative by some evaluator, it must have a substantial degree of originality. However, novelty alone is not enough: closely related to creativity is also the notion of value. A random sequence of symbols may be quite novel but it will not be accepted as creative if it doesn’t have meaning in some shared, accepted language. In her framework for interpreting creativity, Boden (1990) distinguished between two forms of creativity: exploratory and transformational (ecreativity and t-creativity, for short) . Wiggins (2001) proposed a formal way of characterising creativity that accounts for this distinction. In rough terms, ecreativity, the most common type, may be described as a search in a conceptual space C, constrained by rules R, using a search strategy codified by rules T. C is typically very convoluted, and some of its points are difficult to reach by regular search strategies; hence, a particular T will be more successful in producing novel solutions if it is able to reach such points. Achieved concepts will be valued using another set of rules E. While e-creativity is described as an exploration of a conceptual space, t-creativity refers to the transformation of the conceptual space itself: great creative breakthroughs like paradigm shifts fit into this class. Under Wiggins’ formalisation, transforming

C implies changing the rules so that points outside C become reachable. This may be attained by changing one of the rule-sets R or T, which results in a new conceptual space C’. In other words, tcreativity may be described as an exploration in the meta-space of rule-sets. The Creative Systems Group at the University of Coimbra has devoted itself for years to the study of and experimentation with computer models of creativity. Much of the work so far has focused on e-creativity, but, as we will see at the end of this article, recent developments point to even more ambitious goals. Computer models of creativity As stated in many writings in the area of cognitive psychology, the creative faculties of the human mind are highly correlated to the ability to search through spaces or “viewpoints” that are different from the ones immediately involved. For example, according to Marin and de la Torre (1991), our capacities for abstraction, symbolic analysis and finding not-so-obvious relations are associated with creative production. One cognitive psychology theory (Guilford, 1967) concentrates on the idea of “divergent production”. In computational terms, exploring convoluted spaces with the aim of getting to points that are difficult to reach requires flexible search mechanisms, preferably with the possibility of searching disparate areas of the search space in parallel, finding the local maximum without getting locked into it, and diverging as needed. So flexibility is a key issue, and it is for knowledge representation as well. In our own work we have, from our first experiments, used tree-like structures for representing knowledge, and adopted mechanisms that process those structures by reassembling knowledge fragments into novel combinations (Cardoso at al. (2000)). For instance, SICOM (a Music Composition System) resorts to Case-Based Reasoning to construct hierarchical descriptions of musical pieces — much in the same way as in Lerdhal & Jackendoff (1983), where an entire piece of music can be represented analytically by means of a hierarchical structure defined by grouping, time-span and metric rules.

Figure 1: A Music Structure – a structured case in SICOM

The SICOM structures were built in a top-down, iterative sequence. The system used pre-elaborated analysis of music coded as trees, with nonhierarchical links between nodes for establishing relations among them. In the act of producing new structures, SICOM used these links as “suggestions”, with a “strength weight” associated to search space reduction to keep some coherence throughout a piece (for example, in figure 1 Repetition may be strong and Transposition may be weak). IM-Recide, CREATOR and MuzaCazUza are other examples of case-based experiments conducted in the domains of Design and Music, and were all inspired by human models of creativity. But we have also looked for other sources of inspiration. One of these is the Neo-Darwinist theory, which revises Darwin’s first ideas in the light of modern genetics and gives us a scientific framework that explains how life forms survive by adapting themselves to environmental changes. At the core of this process is a mechanism that selects the “fittest” individuals and recombines their genetic material. Putting together “good” parts of different individuals can give rise to a new and better one.This is clearly a way of producing innovative solutions (Goldberg (1998).

21

NEvAr NEvAr (Neuro Evolutionary Art) adopted Genetic Programming as the search mechanism for exploring a conceptual space of images. NEvAr is an Evolutionary Art tool, i.e. a program that allows the evolution of a set of images, based on the aesthetic preferences of the user. NEvAr follows an evolutionary paradigm; in other words, it tries to mimic the mechanisms underlying natural selection, namely: survival of the fittest, recombination of their genetic material, and slight and random modification (mutation). In its basic form, NEvAr operates as follows: a) the program generates a random population of images; b) the user evaluates the images, assigning a “fitness value” to them; c) the program “breeds” a new population of images trough the recombination and mutation of the genetic code of the images of the current population; images with higher fitness values have higher probabilities of being selected for breeding; d) return to point b).

magazine

Figure 2: On the left, the expression f(x)=(x+y)/2 represented in tree format. In the middle, a 3dgraph of the mathematical expression. On the right, an image generated by assigning a greyscale value to each f(x) value.

In NEvAr, the characteristics of the individuals (images) are determined by their genetic code. So we have a phenotype (the individual) and a genotype (the genetic code that, once expressed, results in the individual). The genotypes are trees constructed from a lexicon of functions and terminals. The function set is composed mainly of simple functions such as arithmetic, trigonometric and logic operations.The terminal set is composed of a set of variables x and y and random constants. The phenotype is generated by evaluating the genotype for each (x,y) pair belonging to the image. Thus, the images generated by NEvAr can be seen as graphical portrayals of mathematical expressions (see Figure 2).

Figure 3: An example of the recombination operation.The code of the individuals A and B is recombined by exchanging the sub-trees implicitly defined by two randomly chosen points PA and PB, giving rise to the individuals A’ and B’.

22

As shown in Figure 3, genetic operations (recombination and mutation) are performed at the genotype level. In order to produce colour images we resort to a special kind of terminal that returns a different random value depending on the colour channel – Red, Green or Blue – being processed. NEvAr thus follows an iterative process: as the population number increases the average quality of the images also tends to increase, giving rise to new, interesting, and potentially creative and beautiful images (at least in the eye of the person conducting the program). Figures 4 and 5 give some examples of images generated with NEvAr. One of the misconceptions about evolutionary art tools is that the generation capabilities of a system are deeply connected with the used primitives. Our experience with NEvAr shows that this is wrong. What is necessary, however, is a set of “basic” primitives that can be combined in a powerful way. From an artistic point of view, we consider NEvAr to be a tool with great potential. The generation of an idea results from an evolutionary process and from the interaction between the artist and the tool. Thus, the use of NEvAr implies a change to the artistic and creative process. But in spite of this, the artworks obey the aesthetic and artistic principles of the artist, who guides the process by providing fitness values to the produced images.

Figure 5: Additional images evolved by NEvAr under the guidance of its author.

Figure 6:The author of NEvAr (F. Penousal Machado) working with the tool.

Figure 4: Images evolved by NEvAr under the guidance of its author.

23

Although NEvAr was originally intended as a tool to help people be creative, we are currently studying ways of giving autonomy to the program by automating the fitness assignment. Our initial idea was to train a neural network and use it to automate this task. But we now feel that full automation is not attainable in the short term, and our current idea is to use neural networks (as well as other techniques) as a filter that eliminates undesirable individuals. Figure 7 shows two images generated by NEvAr without any kind of human intervention. Further thoughts If there is one single ultimate goal in computational creativity research, it is undoubtedly t-creativity, for it subsumes the ability to reason at the meta-level, to change the world, and to create new ideas. Pursuing it may seem like a quest for the Holy Grail, yet researching t-creativity forces us to focus on issues that we believe can move us forward to points such as cross-domain transfer processes like analogy, metaphor and conceptual blending.

magazine

Figure 7: Examples of images evolved by NEvAr without human intervention by making automatic fitness assignment..

These cognitive devices motivate the Dr. Divago project, a framework for the generation of new concepts in a multi-domain environment. It uses structure-matching procedures to find candidate mappings for blending parts of (apparently) distant spaces. For example, blending the domains “house” and “boat” (as suggested in Goguen (1999)), Dr. Divago concludes that “the hatch is the window of the boat” or “the mast is the roof of the house”. A drawing module based on logo language coding of elements produced examples such as the ones shown in Figure 8. Conceptual blending allows for the exploration and creation of an alternate, blended domain (e.g.“houseboat”), and therefore makes leaps to unexpected, potentially creative solutions. In Wiggins’ terms, this corresponds to a transformation of the conceptual space by changing the set of rules R. And in this sense, t-creativity is theoretically achievable. Footnote 1 See also Wiggins’ article on pp 7 – 11 of this issue.

Figure 8: At the top, house domain and boat domain instances. At the bottom, two examples of blends generated by Dr. Divago.

24

References Boden, M. (1990) The Creative Mind: Myths and Mechanisms. London: Abacus. Brown, R. T. (1989) Creativity: what to are we to measure? In Glover, J.; Ronning, R. and Reynolds, C., Handbook of Creativity, Plenum Press, New York. Cardoso, A., Costa, E., Machado, P., Pereira, F. C., and Gomes, P. (2000). An Architecture for Hybrid Creative Reasoning. In Pal, S. K., Dillon, T. and Yeung, D. (Eds.), Soft Computing in Case Based Reasoning, Springer-Verlag. Goguen, J. (1999). An Introduction to Algebraic Semiotics, with Applications to User Interface Design. In Nehaniv , C. (ed.), Computation for Metaphor, Analogy and Agents. LNAI, volume 1562, pages 242-291, Springer-Verlag. Goldberg, D. (1998). The design of innovation: lessons from genetic algorithms, lessons for the real world. IlliGAL Report Nº 98004, Department of General Engineering, University of Illinois at Urbana-Champain. Guilford, J.P. (1967) The Nature of Human Intelligence. New York: McGraw-Hill. Marin, R. and de la Torre, S. (1991). Manual de la Creatividad. Barcelona:Vicen Vives. Lerdhal, F. and Jackendoff, R. (1983) A Generative Theory of Tonal Music. Cambridge, Mass.: MIT Press. Wallas, G. (1926). The art of thought. New York: Harcourt Brace. Wiggins, G. (2001) Towards a more precise characterisation of creativity in AI. In Bento, C. and Cardoso, A., ed., Proceedings of the ICCBR’01 Workshop on Creative Systems. Washington, DC: Naval Research Laboratory.

Relevant URLs Creative Systems Group web page: http://creative-systems.dei.uc.pt

Amílcar Cardoso works as an Assistant Professor at the Department of Informatics Engineering, University of Coimbra, where he develops research work as a member of the IALab and coordinator of the Creative Systems Group. He is also chairman of the Working Group on Computational Creativity of the EC - funded COST Action 282. His main research interest is the study and implementation of computer models of creativity. He has a background in Artificial Intelligence, composes music and has experience in industrial software projects. Home page: http://www.dei.uc.pt/~amilcar Carlos Bento works as an Assistant Professor at the Department of Informatics Engineering, University of Coimbra and is a member of the IALab and of the Creative Systems Group. His background is in Artificial Intelligence and experience in industrial software projects. Home page: http://www.dei.uc.pt/~bento Penousal Machado is an Assistant Lecturer at the Department of Informatics and Systems Engineering, Polytechnic Institute of Coimbra, and a member of the IALab, and of the Creative Systems Group, University of Coimbra. His main research interest is the study and development of evolutionary approaches to creativity, particularly the implementation of constructed artists. He is the author of NEvAr.

SICOM project page: http://creative-systems.dei.uc.pt/SICOM.html

Home page: http://www.dei.uc.pt/~machado

IM-Recide project page: http://eden.dei.uc.pt/~pgomes/imrecide/intro.html

ReBuilder project page: http://www.rebuilder.com

Francisco C. Pereira works as an Assistant Lecturer at the Department of Informatics Engineering, University of Coimbra, and a member of the IALab and of the Creative Systems Group. His main research interest is the study and development of computational creativity mechanisms, particularly with respect to blending and metaphor. He has background in Artificial Intelligence.

Dr. Divago project page: http://creative-systems.dei.uc.pt/DrDivago.html

Home page: http://www.dei.uc.pt/~camara

NEvAr project page: http://creative-systems.dei.uc.pt/NEvAr.html CREATOR project page: http://eden.dei.uc.pt/~pgomes/creator/creator.htm

25

Feature

Mon Meilleur AmI Or: can the Information Society really be friendly? Institute for Prospective Technological Studies Seville [email protected] Ken Ducatel

Marc Bogdanowicz

Fabiana Scapolo

Jos Leijten

Jean-Claude Burgelman

The concept of Ambient Intelligence will probably not be totally new to i3magazine readers. Ambient Intelligence offers a vision of the Information Society which emphasizes user-friendliness, efficient service support and support of human interaction. Ambient Intelligence is a response to the way in which Information and Communication Technologies (ICT) have developed recently, with computerised devices becoming more and more ubiquitous, connected and communicating. The idea behind Ambient Intelligence is that if, as seems inevitable, we are going to be increasingly surrounded by such devices, then for the health, comfort and sanity of human society we had better develop intelligent intuitive interfaces capable of recognising and responding to human needs in a seamless, unobtrusive and often invisible way.

magazine

Hence the abbreviation of Ambient Intelligence as AmI, to signal a move beyond concepts such as ‘userfriendliness’, which tends to objectify the relationship between people (as ‘users’) and technologies.AmI should be based on a more seamless and humanistic notion of a people-friendly information society. First some background. The term ‘Ambient Intelligence’ emerged from the work of the Advisory Group to the European Community’s Information Society Technologies Programme (ISTAG). The initial paper, which laid out the concept, aimed to give a strategic orientation to the annual work agendas of the IST programme. But the group wanted their work to have life and impact beyond the short-term horizon of the following two or three years. As a result, in May 2000 the Institute for Prospective Technological Studies (IPTS) was commissioned to prepare a number of scenarios that would provide a ten-year forward look for Ambient Intelligence, as part of ISTAG’s input into the preparation of the next Framework Programme (2002-2006). The project was carried out by an ISTAG working group, chaired by Martin Schuurmans (CEO of Philips Industrial Research), as a collaboration between DG Information Society and the IPTS. Over the ensuing seven months 35 experts from industry, academia and policy organisations, including

26

the European Commission, were engaged in several rounds of scenario development, to which they brought skills ranging from creative and societal analysis to technological expertise. Four AmI Scenarios Four scenarios were built to provide ‘food for thought’ about longer-term developments in Information and Communication Technologies (ICTs). These scenarios give short descriptions of what life with Ambient Intelligence could be like, and give a feel of where the technologies involved could be going; but their main purpose was to provide insights into the technological, economic and social factors that surround the development of AmI (see figure 1). In the rest of this section we give the opening paragraph of each scenario, as well as a brief discussion of the full scenario and the issues it raises.

“Efficient” 3: Carmen: traffic optimisation

1: Maria: personal ambient communicator

Community

Individual 2: Dimitros: connecting people and expressing identities “Sociable, humanistic”

4: Annette and Solomon: social learning by connecting people and creating a community memory

Figure 1: Four Ambient Intelligence Scenarios.The two axes and the four scenarios provide a structure representing the main features and alternative development paths for Ambient Intelligence.

Scenario 1: Maria Personal ambient communicator After a tiring long-haul flight Maria passes through the arrivals hall of an airport in a Far Eastern country. She is travelling light, hand baggage only. When she comes to this particular country she knows that she can travel much lighter than in those days less than a decade ago, when she had to carry a collection of different so-called personal computing devices (laptop PC, mobile phone, electronic organisers and sometimes beamers and printers). Her computing system for this trip is reduced to one highly personalised communications device, her ‘P–Com’, which she wears on her wrist…. One possible defining feature of AmI is its orientation towards an élite business clientele.The “Maria” scenario provides a model of this potential development. This type of scenario, which basically extrapolates from the mobile business market, is likely to have a high profit potential, because of the high disposable income business travellers tend to have, and the market pull of the companies that they work for. It may incorporate a series of inter-operating ambient intelligence pools (airports, hotels, conference centres, automobiles, smart highway…), since business travellers require a lot of related services (such as hotel reservations and car hire). AmI is likely to be strongly service-based, and service providers are already in the vanguard of applications of new ICTs in order to manage client accounts (with loyalty programmes and so on). From the human perspective, however, the “Maria” scenario leaves something to be desired.: as the rest of the scenario reveals, Maria is rather isolated and seems to live and work in some kind of “pressure cooker”. Her responses to her possibly stress-related health problems are not necessarily the most appropriate.

27

Scenario 2: Dimitrios Connecting people and expressing identities It is four o’clock in the afternoon. Dimitrios, a 32- year-old employee of a major food-multinational, is having a coffee at his office’s cafeteria, together with his boss and some colleagues. He doesn’t want to be excessively bothered during this break. Nevertheless, he is receiving and dealing with incoming calls and mails all the time. He is proud of ‘being in communication with mankind’, as are many of his friends and some colleagues. Dimitrios is wearing, embedded in his clothes, a ‘gateway’ or digital avatar of himself, familiarly known as ‘Digital Me’ or ‘D-Me’. A D-Me is both a learning device, learning about Dimitrios from his interactions with his environment, and an acting device, offering communication, processing and decision-making functionality. Dimitrios has partly ‘programmed’ it himself, at a very initial stage: it was, he says, a great personal experience to formalise his identity and the way he envisaged his relations. At the time, he thought he would ‘upgrade’ this initial data periodically. But he didn’t. He feels quite confident with his D-Me and relies upon its ‘intelligent PDA-like’ reactions… Maintaining existing relationships and creating new ones is an essential feature of human life, and a major driver when it comes to telecommunication revenues (mobile, emails): ‘communicating’ has obvious mass consumption potential.This means that there is a growing demand for targeted forms of communication and relationships on the one hand, and multiplied inter-communication on the other.The D-Me is a people-based, ad-hoc networking device which registers, processes and offers information on private lives. Its purpose is to facilitate social relations and networking, provide communication interfaces and take decisions on behalf of the wearer in specific situations. The Dimitrios scenario is closer on the time horizon than scenarios 3 and 4. The emphasis is on play and social interaction rather than on ‘efficiency’, and it may therefore appeal particularly to ‘alternative’ or ‘youth’ cultures, which could be among the first to emerge as leading markets. The scenario assumes a change in people’s behaviour, mainly in terms of their willingness to reveal (or disguise) their personality on-line. Price could be a barrier to a break-through in the mass market.

magazine Scenario 3: Carmen Traffic optimisation It is a normal weekday morning. Carmen wakes and plans her travel for the day. She wants to leave for work in half an hour and asks AmI, by means of a voice command, to find a vehicle to share with somebody on her route to work. AmI starts searching the trip database and, after checking the willingness of the driver, finds someone who will pass by in 40 minutes.The in-vehicle bio-sensor has recognised that this driver is a non-smoker – one of Carmen’s requirements for trip-sharing. From that moment on, Carmen and her driver are in permanent contact if they want to be (e.g. to allow the driver to alert Carmen if he/she will be late). Both wear their personal area networks (PAN) allowing seamless and intuitive contact… The “Carmen” scenario is further out on the time horizon than the two scenarios sketched above, not so much because of technological barriers but because it implies major infra-structural developments (namely highly-developed networks of inter-operating sensor systems and dynamic database management systems). It describes an ambient landscape in which the joint flows of bits and atoms are optimised to create a more sustainable urban system. This scenario makes significant assumptions about changes in public behaviour, such as accepting ride shares and traffic management systems. It re-conceptualises the functioning of the transport network, with distribution of goods based on real-time demand. It presupposes that inter-modality is very efficient for the transport of both people and goods and that, at governmental level, legislation is in place to allow the city to work as an organisational system chain.To ensure sustainability it uses energy optimally and maximises safety.

28

Scenario 4: Annette and Solomon Social learning by connecting people and creating a community memory It is the plenary meeting of an environmental studies group at a local ‘Ambient for Social Learning’.The meeting is led by a human mentor who guides and facilitates the group’s operation but who is not necessarily very knowledgeable about the subject of environmental management. The plenary ‘meeting’ takes place in a room looking much like a hotel foyer, with comfortable furniture pleasantly arranged… A member arrives late.As she enters the room and finds herself a place to work, she hears a familiar voice asking “Hello Annette, I got the assignment you did last night from home: are you satisfied with the results?” … The voice suggests that Annette spends some time today trying to pin down the problem with the model, using the new enhanced interactive simulation and projection facilities. It then asks her to give a brief to the group and goes briefly through its understanding of Annette’s availability and preferences. Finally, Annette agrees on her work programme for that day. The ambient schedules a viewing for some of the other members who might be interested, but cannot be present…. This scenario describes a learning environment — the Ambient for Social Learning (ASL). Its starting point is that learning is essentially a social process.The ASL environment supports and upgrades the roles of all the actors in the learning process, starting with the mentor and the students. It is capable of creating challenging interacting learning situations, co-designed by the mentor and students in real time. Students using ASL are producers of learning material and create input for the ‘learning situations’ of others. In other words, the ASL is both an environment for generating new knowledge for learning and a place for learning about learning. The ASL is also a physical space (a room or a group of rooms) and its ambient facilities include many links with similar places. Its layout and furnishings are flexible and diverse, so that it can serve the learning purposes of many different kinds of groups and individuals. The Carmen scenario is probably the furthest out in terms of time as it makes high demands, both from a technological and socio-economic viewpoint. It implies significant technical developments, such as high “emotional bandwidth” for shared presence and visualisation technologies, and breakthroughs in computersupported pedagogical techniques.The scenario also presents a challenging social vision of AmI in the service of fostering community life through shared interests.

Social and political factors Working with the four scenarios made it very obvious that AmI raises some very serious social and political challenges. If it is to achieve its ambitions, or even if it is to be accepted by society, it should be driven by humanistic concerns. All experts agreed on this requirement, irrespective of their background. What was less clear was whether we will succeed in meeting these humanistic concerns in AmI. For example, the scenario of Dimitrios, with his rather frenetic multi-tasking communication-intense lifestyle, carries hints of information overload and not daring to disconnect. Issues of information governance were also high on the list of concerns of the scenario group. It was made clear that AmI should be constructed in the framework of a design philosophy that protects privacy and individual choice, because it could too easily be perceived as a ‘Big Brother’ surveillance and control tool.. There are, of course, technological

ways of protecting privacy and security. Personal agents could have the capacity to recognise different levels of exposure and to hide or reveal personal information using cues about the risk level of the transaction, the level of intimacy permitted to the respondent and so on. It is possible nowadays to hang virtual ‘do not disturb’ signs on our mobile phones and email boxes through the use of filters, answering machines and judiciously used on/off switches. But these often rather clumsy strategies will need to evolve into more sophisticated forms. AmI is social technology, but is it also convivial? The Dimitrios scenario shows Dimitrios operating in a social context, as evidenced, in the full version of the scenario, by his virtual assistance to an elderly gentleman. But the full version also shows that Dimitrios’ wife is not entirely happy with the way he uses his D-me to play hide-and-seek with his family responsibilities. This juggling of responsibilities and priorities has always been a feature of daily life, but being accessible all the time makes us much more

29

exposed. It is clear that AmI could be used as a means of social subterfuge and deceit as well as increased transparency. The scenario of Solomon and Annette is firmly embedded in the social context of learning, and responses to the report indicate that it is the scenario that readers find hardest to accept and understand. This is not so much because it presupposes sophisticated technologies to manipulate the virtual environment, but because it requires significant advances in the understanding of human knowledge, interaction and behaviour. It does not supplant human (inter)actions; it assists them. Not all of our readers were willing to believe that this path of development was possible. Overall, the experts were most concerned with envisioning a sustainable AmI world. What new health risks (e.g. from magnetic radiation) would we face? What new digital divisions might emerge based on income, education, skills and age? Will the hypermobile AmI-enhanced future be one in which the use of resources is profligate? Business and industrial models

magazine

Economic questions abounded in the scenario group. Who will make AmI work? Who will produce it? Who will live and consume in a world like that? Who will make money? Which businesses will grow and which will decline? How will AmI change the way people work, think, learn, and communicate, and how will it change the settings in which they do this? The sheer diversity of possible factors defeats the establishment of a common economic baseline with major certainties and uncertainties. However, a number of elements emerged from the scenario work as important for inclusion in the tenyear timeframe. Five main drivers of demand for AmI were rated as important, namely: • improving quality of life (including satisfying ‘intangible’ needs such as better community life and health, as well as rising material demands); • enhancing productivity and quality of products, services, and applications; • developing new products and services for new and emerging AmI firms; • developing new AmI applications in industrial innovation and new products (e.g. household and office equipment, clothes, furniture..); and • developing applications in public services such as hospitals, schools, police or the military.

30

The technologies Finally, we turn to the technological underpinnings of the scenarios. In the course of the scenario-building work a wide range of technologies were identified as key enablers of AmI. On the basis of the scenarios it was possible to develop a set of common technological requirement areas. Due to the sheer diversity of technologies involved neither this list nor the terminology can be definitive, but the list does suggest which technologies will be essential for AmI and which technological trajectories they imply. Requirement 1: Very unobtrusive hardware Miniaturisation is assumed to follow its historic pattern to permit the necessary enabling developments in micro, nano and optical electronics. Key hardware features could be: selfgenerating power and micro-power usage in objects such as very low-power radio-frequency chips; new forms of input/output smart surfaces, paints and films that have smart properties; proliferating active devices such as sensors and actuators, integrated with interface systems in order to respond to user senses; posture and environment or smart materials that can change their characteristics; and performance by stand alone intelligence or by networked interaction (e.g. smart clothing). A key feature would be the development of a human-factors design emphasis, so that the widespread embedding of computers produces a coherent AmI landscape rather than just a proliferation of electronic devices with IP addresses. Requirement 2: A seamless mobile/fixed web-based communications infrastructure Complex heterogeneous networks need to function and communicate in a seamless and interoperable way. This implies a complete integration (from the point of view of the user or network device) of mobile and fixed radio and wired networks. Probably all the networks would be operating with some equivalent of IP technology. Networks will have to be dynamically reconfigurable, requiring more advanced techniques for dynamic network management. Requirement 3: Dynamic and massively distributed device networks The AmI landscape contains almost countless inter-operating devices: wired, wireless, mobile and fixed. Databases, whether centralised or

distributed, should be accessible on demand from anywhere in the system. This complexity extends well beyond the current capabilities of system software and middleware, and calls for wireless ‘Plug and Play’ solutions as well as dynamic, multidomain networking. The way to construct these networks is a major research challenge for the coming years.

A host of new business models will emerge, no matter what the response by Europe. These new business models will be tested by industry and entrepreneurs operating in the areas that provide the most fertile conditions for experimentation.The vision of Ambient Intelligence points at how to create these fertile conditions, in the technological domain and in business environments.

Requirement 4: A natural-feeling human interface

But all the scenarios and all the experts involved emphasised the social dimension of innovation, the ability as well as the willingness of society to use, absorb or adapt to technological opportunities. Issues such as environmental and social sustainability, privacy, social robustness and fault tolerance will determine the take-up of AmI. Thus, although technological feasibility is fundamental and an essential precondition, our work on AmI underscores the requirement that policies for the information society should be driven by the usefulness and relevance of its applications, rather than by enthusiasm for technologies per se.

A central challenge for AmI is the creation of systems that are intuitive to use – almost like human functions such as breathing, talking or walking. This means that, on the one hand, ‘artificial intelligence’ techniques will have to be employed for dialoguebased and goal -orientated negotiation systems, as the basis for intelligent agents and real time middleware. The key issue will be to move from relatively narrow domain-by-domain and highly structured databases to families of systems that can operate across domains to very general levels. Requirement 5: Dependability and security A consistent theme running through the scenario work was the need for a safe, dependable and secure AmI-world. The scenarios assume techniques for secure ID authentication, micropayment systems and biometrics. These sorts of ‘trust technologies’ and advanced encryption techniques are strong requirements. Conclusions The scenarios underscored the importance of wellfocussed and world-class long-range ICT research in Europe. There is an urgent need to open up the growing reservoir of new ideas and creativity to help invent the future. This implies that whatever measures are taken in the Framework Programmes to support Europe’s ICT development, they should support fast responses, flexibility and world-wide scientific and technological co-operation, as well as entrepreneurship.

The views expressed in the article are those of the authors and do not necessarily reflect those of the European Commission.

The authors are (or have recently been) members of the Futures Project and the ICT unit in the Joint Research Centre’s Institute for Prospective Technological Studies of the European Commission. Jos Leyten rejoined TNO-STB in the Netherlands in May 2001. For further information on ISTAG, their recommendations & reports, see http://www.cordis.lu/ist/istag.htm The full report is available on-line: K. Ducatel, M. Bogdanowicz, F. Scapolo, J. Leijten & J-C. Burgelman, Scenarios for Ambient Intelligence in 2010 (February 2001): http://futures.jrc.es.

The scenario exercise showed that the vision of Ambient Intelligence is a strong starting point for giving direction to research. Major opportunities to create an integrated Ambient Intelligence landscape can be built upon European technological strengths in areas such as mobile communications, portable devices, systems integration, embedded computing, multi-platform content provision and intelligent systems design.

31

Feature

Computer games – for better or for worse? Ghostwriter Judy Robertson, University of Edinburgh. [email protected]

Computer games are a part of many children’s lives. While some parents and pressure groups are concerned that children may become addicted to computer games or mimic the violent behaviour depicted in some games, researchers within the i3 community and beyond are exploring the potential educational benefits of computer game technology.The integration of motivating graphics and audio with the narrative components of compelling plot and characters can give children challenging and enriching virtual adventures. In particular, educational drama exercises such as role-play can take exciting new forms within a virtual environment, teaching children decision-making, empathy and negotiation skills as part of a game. In this article Judy Robertson describes Ghostwriter, an example of a virtual environment for role-play activities, and discusses children’s experiences with it.

Children’s exposure to computer games

magazine

Many children enjoy playing computer games. In a recent field study (Robertson, 2000), I asked 30 tenyear-old pupils how frequently they play computer games and how long they spend playing them at a time. Although the children found it difficult to judge this, the most common length of time spent at one session was half an hour, and the most common frequency of game sessions was three to four times a week.A few children played games every day of the week, for “hours at a time”. Some of the children stated that their parents restricted their use of games because “they had been playing them too much”. Indeed, adults’ concern about children’s “addiction” to computer games has provoked what McNamee (1998) terms a “moral panic”. The pleasure which children take from computer games, the amount of time which they devote to playing them, and the violent themes of some games has led to concern about the impact they could have on children’s social behaviour. But can children’s fascination with computer games technology be harnessed for educational goals? The answer appears to be “yes”. Recently, several research projects, some of which funded by i3 ESE, have begun to explore the possibilities of using virtual-environment and computer-games technology for children’s education. the POGO project (Rizzo and Saudelli, 1999), the NIMIS project (Machado, Martinho, and Paiva, 1999) and the Puppet project (Klesen et al., 2000) are examples of this. Unlike commercial computer games, which are notorious for extremely flimsy story-lines, these projects integrate motivating virtual-environment technology with narrative.

32

Commercial computer games can suffer from the problem that they encourage children to act repeatedly (and often violently) without thinking and without concern for the consequences of their actions.They also offer few opportunities to interact in any meaningful way with other people. But stories can provide a heart for the otherwise empty shell of a virtual environment, and can engage children’s emotions while posing difficult moral decisions. A compelling plot and the concerns of a cast of characters can elevate a game above a simple problem-solving or object-harvesting romp through a virtual world. Ghostwriter, a virtual role-play environment in which children take part in a story, illustrates this. Ghostwriter Ghostwriter is an adaptation of the commercial computer game Unreal by Epic Megagames, Targetted to 10-to-12-year-olds, it was designed for educational role-play and story-writing activities during my PhD work in Edinburgh (Robertson 2000; Robertson and Oberlander 2001).The goal of the PhD work was to establish the effect of a narrative virtual environment on children’s learning; both on their writing skills and on their social and personal development, which form a major strand of the Scottish National Curriculum. Evaluation of the effect of Ghostwriter on characterisation in children’s writing has shown that playing the game encourages children to write about the relationships between characters (Robertson, 2000). But in this article, I’d like to focus on the aims of the role-play activity itself, which are to encourage children to empathise, negotiate and think carefully about decisions. These skills are targeted in the

personal and social development strand of the Scottish National Curriculum, and although difficult to approach through direct instruction, they can be fostered through drama activities. During the virtual drama, children play the parts of a brother and sister named Jenny and Daniel.They have been asked by their Grandmother to find her old friend Fred, who lives in the castle on Kirkmystery Hill. When they enter the virtual castle, the children meet Fred but discover that he cannot leave the castle because he is trying to protect it from an evil force.An evil sorceress, named Lady Searle, is trying to take over the castle. Unknown to Fred, Lady Searle has already infiltrated the castle, disguised as a book. Once the children encounter the book, the outcome of the plot is determined by their actions. Ultimately, the children must decide whether to kill Lady Searle to prevent her taking control of the castle. Fred and Lady Searle are both played by the role-play leader. In the Ghostwriter environment, two children and one (human) role-play leader interact with each other in the virtual world. Each role-player controls an avatar in order to move around this graphical world, and improvises dialogue for their story character. In the original version of Ghostwriter these improvisations were typed messages (Robertson, 2000), but as an extension to the original game these were replaced with voice communication using walkie-talkies. With the help of a trained actress and an experienced teacher, I piloted the walkie-talkie version of Ghostwriter with 30 10 and 11-year-olds in a state funded primary school. Children’s responses Interviews with the children afterwards showed that the children found the game extremely motivating: “It was really, really fun!”;“It was exciting and you could communicate with people”; and “It was the best computer game ever”.The children also offered some insightful comments about the benefits of the game: “It teaches you how to communicate. It could be helpful to children because if they were shy it would tempt them to talk.”; “It was mind-challenging and you had to make hard decisions.”;“It had a good storyline”;“It felt as if I actually was the character and things were really happening to me.”. The children were also able to reflect on and explain their own thoughts and behaviour during the drama

Figure 1: A scene from Ghostwriter. Lady Searle tries to persuade Jenny to join her cause

(“She sounded friendly and nice, and I really did trust her. She kept talking me into things, and then I thought she was evil.”) and empathise with the characters in the story . In answer to the question “How do you think Lady Searle felt during the game?”, the children suggested that “She thought we were stupid”; “She was just pretending to be nice to try to catch you”; “I would hate to be trapped in the book”;“She must have been happy when she got out of the book”. As they played the game, the children appeared to suspend disbelief and become involved in the adventure. This is apparent from the volume, pitch and speed of their voices in the recordings (you can listen to an excerpt at www.cogsci.ed.ac.uk/ ~judyr/ghostwriter/voice/example1.mp3) I noticed that the children seemed to think more carefully about their decisions when using the walkie-talkies than when typing messages. As an illustration, one decision which the children must make early on is whether to go with the book to find Fred. Fred has asked the children to stay and guard the amphitheatre while he goes to fight the evil Lady Searle. But Lady Searle, disguised as a book, tries to persuade them to leave the amphitheatre, pretending she will take them to Fred. So the children have to decide whether to break their promise to Fred.The following excerpt illustrates the issues surrounding this decision:

33

Conclusion Evaluations of Ghostwriter so far show that children love to play the game.They enjoy the chance to use their game skills at school, and often prefer the communication and character interactions in Ghostwriter to the violence of the computer games they play at home. But more importantly, as the children play they are practicing important social and personal skills, such as perspective-taking, negotiation and decision-making. Or, as the teacher who took part in the study remarked: “You can teach them all kinds of useful things without them even noticing!”. References

Figure 2: Fred and his dog, on guard in the amphitheatre Lady Searle: How about we go on a mission? Do you both think we could? Jenny: I’m not too sure.

Machado, I., Martinho, C., and Paiva,A. (1999). Once upon a time. Published in Fall Symposium on Narrative Intelligence of AAAI 1999.

Lady Searle: Not sure? Still not sure. I am Fred’s friend. I don’t know how to prove it to you because I’m stuck in this book.

McNamee, S. (1999). Computer and video games: special objects or everyday artefacts in children’s worlds? Presented at European Sociological Association 1999 Will Europe Work Conference, Stream V.2. – Sociology of Childhood.

Jenny: We’ve been told to stay here.

magazine

Klesen, M., Szatkowski, J. and Lehmann, N. (2000). The Black Sheep Interactive Improvisation in a 3D Virtual World. In M. Caenepeel e.a. (eds). Building tomorrow today Proceedings of the i3 Annual Conference 2000.

Lady Searle: But the quickest way to get out of this castle is to find Fred and I think I know where he’ll be. Why do you need Fred anyway? Daniel: Our grandmother’s lonely. She wanted to find Fred. Lady Searle: That’s so sad! Well we’ve got to help. It’s even more important then. Jenny: We made a promise.You can’t break promises. Lady Searle: I know and I think you’re very, very wise to say such a thing. But the thing is, Fred can’t do it by himself. He’s old. He thinks he can. But I think we could help him.

Rizzo, A., and Saudelli, B. (1999). POGO. Pedagogical issues in designing narrative technology. In: M. Caenepeel e.a.: Community of the Future. Proceedings of i3 Annual Conference, 1999. Robertson, J. (2000). The Effectiveness of a Virtual Role-play Environment as a Preparation Activity for Story Writing. PhD Thesis, University of Edinburgh. Robertson, J. and Oberlander, J. (2001). Ghostwriter: Educational drama and presence in a virtual environment”. Paper submitted to Journal of Computer Mediated Communication.

Judy Robertson is a research associate within the Division of Informatics at the University of Edinburgh, with an interest in the development of children’s written and oral narrative skills.The Ghostwriter project was carried out during her recent PhD research, also at the University of Edinburgh. Judy is also a children’s storyteller and is working with the Scottish Storytelling Forum on plans to link the ancient art of oral storytelling with new technologies. Ghostwriter website: www.cogsci.ed.ac.uk/ ~judyr/ghostwriter/index.html

34

Report

Extreme Interfaces Media Lab Europe, Dublin, 28 January 2002 Extreme Interfaces, an event held earlier this year at Dublin’s recently established Media Lab Europe (MLE), was an occasion for presentations and dialogue involving the local research team, collaborating MIT Media Lab researchers and invited visitors. It also provided an opportunity to gain an impression of the directions in which the work of lab is proceeding. One prevailing direction at MLE could be characterised as human-centredness.This is evidenced in a broad range of new research projects in which human expression, sense perception and touch hold centre stage, both in computer-mediated human-tohuman communication and in the study of human interaction with advanced communication systems. Sense perception as a research focus is reflected in project titles such as palpable machines, tangible bits, haptic suits and affective systems. Palpable Machines, a research group led by Sile O’Modhrain, embraces a cluster of related studies involving haptic displays and force-feedback devices. The group’s goals include user performance enhancement in the manipulation of remotely controlled machines, as well as simulation enhancement in the conduct of remote online medical examinations. The human factor also informs Lorna Ross’ work on mobile devices for physical and mental health. Other areas of research focus on the individual citizen and the social community the individual is part of. This is illustrated by the Human Connectedness group, led by Stefan Agamanolis, which addresses a world in which

Patrick Purcell Imperial College London [email protected]

the personal and societal problems associated with distance and isolation may be alleviated with the help of technologies that engender a sense of togetherness among individuals, families and neighbourhoods. Music plays an important and multi-facetted role in the research agenda of MLE. The overarching vision is to exploit today’s sensor and informationprocessing technology in order to augment musical expression on the part of both professional musicians and lay people, with a special emphasis on the potential of children. A prime example of work in this area is the Toy Symphony, composed by Tod Machover (who leads the music research activity at MLE), which was recently given its world premiere in Dublin by the Irish National Orchestra. This new symphonic composition, which is a major vehicle for demonstrating the many facets of music research at MLE, can function as a testbed for a range of specially developed electronic instruments, both to enhance the performance of skilled virtuosi and to encourage children’s learning processes and expressive and creative powers.

Figure 1:The hyperviolin is a recent addition to the hyperinstruments collection, a family of computer-linked instruments specially developed for enhanced virtuoso performance in Tod Machover’s innovative compositions (such as the Toy Symphony, and previously the Brain Opera).

Figure 2: Beat bugs are hand-held percussive instruments which allow both children and adults to shape and modify musical lines with both vigourous expressive gesture and delicate touch.The beat bug was developed initially for performance in works such as the Toy Symphony.

35

Recent instrumental innovations include the hyperviolin (figure 1), the shaper and the beat bug (figure 2). While the hyperviolin’s role extends the performance of the professional musician, the other two instruments were developed primarily for children. The hyperviolin is linked to a symbiotic computer system which measures and interprets the players’ musical expression. Shapers are soft, squeezable devices that allow players to mould, transform and explore musical material. Beat bugs are hand-held percussive instruments that, when linked together in an ensemble of players, can share in, and help develop, rhythmic patterns, from which large-scale collaborative compositions may be formed..

magazine

The increasing availability of sophisticated electronic sensor technology and advanced information processing software is having a significant impact on the variety of electronic music interfaces currently being developed. This upsurge of interest in, and experimentation with, new areas of musical expression inevitably raises questions about the future direction of music. To stimulate discussion of such questions, Media Lab Europe is hosting an international conference at the end of May 2002, with the title NIME 2002 (New Interfaces for Musical Expression). Bringing together academics, technologists and performance artists, this will provide a venue for demonstrating leading international examples of innovation in instrument development. It will also act as a forum for discussing the implications of such developments for contemporary music composition and performance.

One of the more intriguing exhibits at the MLE Extreme Interfaces event was a prototype, aptly called the Haptic Suit (figure 3), which is the outcome of a project by MIT researcher Eric Gunther. While music relies for it appreciation on acute auditory sensibility, Gunther’s project sets out to demonstrate that there may be a radically new and complementary art form, predicated on an equally acute human sense of touch.The haptic suit is in effect a whole-body-enveloping “instrument” that facilitates the perception of musically-structured spatio-temporal patterns of tactile vibration on the body’s surface, with concurrent patterns of audible vibrations as musical accompaniment.The wearer of the Haptic Suit can either be a single individual or a member of a small audience (similarly accoutred in haptic suits!). The Extreme Interfaces programme also included several presentations which extend the range of human-system interfaces into the area of biomedical signal processing. Project demonstrations showed how the computer may alter performance as a result of becoming aware of a change in the affective state of the human subject, and how the human subject may be influenced in turn by the consequent change in the performance of the computer.

Figure 3:The Haptic Suit is an outcome of the Cutaneous Grooves project, whose overall aim is to exploit the potential of the human sense of touch as a delicate sensory channel, capable of appreciating the most refined forms of artistic expression. What is not immediately apparent is that the wearer is experiencing a full-body, vibro-tactile sensation, coupled with accompanying music.The orange suit is lined with vibrating transducers.Various signals are passed through the transducers to create a ‘tactile composition’ which can be orchestrated with the accompanying music.

36

Figure 4: A frame from “Mind Games”, a multiplayer computer game in which the competing players control the performance of flying dragons by means of bio-feedback sensors attached to the individual player’s fingers. As each player starts to relax, the dragons begin to fly forward.The winner is the dragon whose player achieves the greatest degree of relaxation in what may be a competitive and stressful computer game environment.

One such project is Mind Games, directed by Gary McDarby. Mind Games is a competitive multi-player game in which the mood of the individual player affects the pace of the animated computer creature (a flying dragon) through bio-feedback sensors attached the player’s hand. The more effectively the individual player can control her relaxation level, the more effective the performance of the animated screen dragon will be in terms of competing in the race (figure 4). The current research has many application areas, mainly therapeutic ones, such as the treatment of attention disorders in children. The wide-ranging research work demonstrated and discussed at the Extreme Interfaces event was mainly carried out by a team of some 40 research staff at Media Lab Europe; but it also included contributions by collaborating MIT Media Lab researchers, as well

as joint work by MLE projects and local Irish universities. Together this adds up to a growing research programme, currently consisting of more than 20 projects ranging from Creative Music Education, through Wearable Bio-monitors to (a proposal for) A Brain Computer Interface. In just under two years Media Lab Europe has established itself in Georgian Dublin’s historic centre on a prime site of the city’s industrial heritage, and has engaged in a forward-looking programme of research — a programme which, though technology-focused, embodies a real concern with associated social, cultural and human issues. In effect a 21st century research agenda in an 18th century ambience!

Patrick Purcell is Visiting Professor and Senior Research Fellow at the Intelligent and Interactive Systems Group of the Department of Electrical and Electronic Engineering, Imperial College, London. His work reflects a primary and abiding interest in the application of technology to various aspects of human affairs, especially in the application of new media in the design of the user interface. Patrick is a member of the i3 Coordinating Group. URL: http://www.ee.ic.ac.uk/hp/staff/purcell.html For more on Human Connectedness: http://web.media.mit.edu/~stefan/hc/ Mind Games: http://mindgames.mle.ie/ Palpable Machines: http://www.mle.ie/palpable/default.shtml

Figure 5:The MLE open plan area. An interior view of MLE, set in what was previously the Guinness Hop Store.

Elixir: http://www.mle.ie/research/WB/index.html Gunther: [email protected] Media Lab Europe: http://www.mle.ie/

37

Remembering Mike Scaife

Many i3 members have deeply felt the loss of Mike Scaife who died, very suddenly, on 18 December 2001 (aged only 53).Yvonne Rogers, Mike’s partner in life, work and vision, shares some of her memories of Mike, an outstanding researcher and “a very special and great friend, who will be missed beyond words”.

magazine

Those who had the privilege to know and work with Mike Scaife will never forget him. He was always so full of life, overflowing with charm, wit, and enthusiasm. To encounter Mike was to meet a playful, wickedly intelligent spirit; his passion and zest for life was simply infectious. A co-founder of the School of Cognitive and Computing Sciences, at the University of Sussex, Mike was a Reader in the psychology group. He was one of the few researchers who you could genuinely say was an interdisciplinarian and, moreover, was able to reap the benefits of doing so. A continuing thread throughout Mike’s research was the application of cognitive and developmental psychology to problems in cognitive science and humantechnology interaction. He was concerned primarily with the design and implementation of IT to maximize the educational potential of children. His focus was on developing novel forms of interactive technologies to extend and augment current ways of learning and playing. When invited to become part of an i3 ESE consortium to create a virtual puppet theatre for young children to develop their narrative skills, Mike immediately saw how this could be achieved. Based on our theory of external cognition (Scaife and Rogers, 1996)1 , Mike’s idea was to create a virtual environment where children would be able to reflect on what they were doing in ways that moved beyond the everyday ‘here and now’ of acting, by allowing them to take multiple perspectives in the

38

creation of their own plays.To achieve this required us working as a multidisciplinary team, with partners from a variety of backgrounds (cognitive science, developmental psychology, HCI, dramaturgy, virtual environments, and autonomous agents). For many such a challenge can be quite daunting; for Mike it was the opposite, providing him with the scope for masterminding a great vision. From the early days of co-writing the PUPPET proposal to the final stages of seeing the PUPPET system actually being used by children to create narratives in highly innovative ways, Mike was very much at the helm, providing the inspiration, together with the more down-to-earth ‘glue’ to keep all the parts sticking together. Even when what seemed like insurmountable obstacles got in the way (and there were many), Mike always had something up his sleeve to overcome them. His enthusiasm to solve problems and continue to push the research agenda forward were relentless. Disciplinary boundaries and rigid ways of doing things were an anathema to him. While at Sussex, Mike co-founded the Interact Lab, which has become internationally known. A driving question that underpins the research done in the lab concerns the use of various types of representations in problem-solving, creativity and play. A team of research fellows and post-grad students from varied backgrounds continue to work and move the research forwards in ways that Mike would have been proud of (cogs.susx.ac.uk/interact).

Paul Marshall, who joined us recently as a research fellow working on the PUPPET project and who is now doing a PhD in the lab, says:

And, finally, may he continue to wear the badge of being the funniest man in i3. Yvonne Rogers, COGS, Sussex University.

“I only knew Mike for 6 months. However, as another member of our lab perceptively said of Mike, he had a way of making people feel that they knew him better than they actually did. To work with Mike was to share in the obvious passion he had for his research, especially evident when he was talking to the children for whom the PUPPET system was designed, and in his sharp sense of humour. He was always patient and generous with his time, although for Mike, time must have been a very valuable commodity indeed. Mike’s death is tragic, not only for his family and friends, but for the students and colleagues, present and future, who would have benefited from his wisdom and experience.”

Footnote 1 Scaife, M. & Rogers, Y. (1996) External Cognition: How Do Graphical Representations Work? International Journal of HumanComputer Studies, 45, 185-213.

Mike Scaife memorial page and obituary by Margaret Boden: http://www.cogs.susx.ac.uk/users/mikesc/ Mike Scaife home page (which is being kept intact): http://www.cogs.susx.ac.uk/users/mikesc/indexoriginal.html

News Recent developments in i3 i3net received one year’s funding for its next phase, which started on 1 March 2002 and will finish on 28 February 2003. The funding is intended to sustain community-building efforts while a new proposal for a new network supporting the community that has formed around initiatives like i3 and DC is being prepared. Niels Ole Bernsen resigned as i3 coordinator but agreed to act as a ‘caretaker’ coordinator during interim period. Simon Bensasson writes: “I would like to take this opportunity to thank Ole for his very hard and dedicated work in the formative years of the network, whose success is measured by the degree to which it managed to forge a research community where previously there was none. No one can dispute this success and I believe one can say with confidence that i3 is a community without precedent in the world. I think I can safely say that Ole’s contribution to this was substantial and I say so in the knowledge of the difficulties that any new

network faces, needing to balance central drive to get things done with open consultation to make sure that the whole community is represented. I am convinced the community that started forming around the i3 initiative and was supplemented by the subsequent ones such as i3-ESE, DC, and soon the Presence initiative, is a much needed one, and one whose importance cannot but increase in the years to come.The future shape of course will not be constant — new concepts will continue to emerge, as well as technologies that will help such concepts become a reality. For this reason we look forward to seeing a proposal that represents the whole community for a network that is open, appropriately structured, dynamic and, especially, participative. Moreover, I hope this would be a proposal that would not be centred on projects running at any one time (this is would be a very transient reality) but on the community of the emerging domain as a whole.”

39

First i3 Summer School reserved for hands-on ateliers allowing students to work together in groups on smaller projects.

The first i3 summer school took place in late summer 2001 in the lovely North-Italian town of Ivrea, located half-way between Milano and Torino, just at the mouth of the Aosta valley. Lecturer and participant Thomas Rist writes: Ivrea is often associated with the typewriter and office-systems giant Olivetti, who is indeed heavily represented in the town through a large number of office buildings and production facilities; and the venue for the school, the brand-new Interactive Design Institute, was situated among these.

magazine

The motto of the summer school, which was directed by Gillian Crampton Smith and Giorgio De Michelis, was Designing for Communities. In the mornings invited speakers gave lectures on various design-related topics, while the afternoons were

My affiliation with the school was manifold. I assisted somewhat in the preparation (many thanks to Gillian, Giorgio and all the others who did the real work!), gave a lecture on Designing Interfaces with Animated Agents, took part in the Virtual Character Design Atelier, and (last but not least) enjoyed socialising with students and other lecturers, in one of the various local pizza restaurants or at an afterworkshop beach party at nearby lake Sirio. In fact, I can hardly imagine a better opportunity to meet such an international mixture of people from different disciplines all sharing an interest in interaction design. From a teacher’s perspective I am still impressed by the high level of expertise that the students showed in their contributions to panel discussions and atelier work. While some of the students had already heard about i3, the school provided an excellent forum for informing participants about the i3 vision, i3 projects, and the possibility to join the i3 community. If it turns out that the longer-term impact of i3 lies not so much in concrete new products but in bringing about a change in culture, then a summer school like the one we had in Ivrea, is an excellent vehicle to keep this culture alive. Unfortunately there will be no i3 Summer School this year, as the EC was unable to fund it.

Convivio 30 i3 and DC members met in Brussels on 11 January 2002 to discuss and contribute to the development of a new network proposal. Topics of discussion included the network’s aims, vision, role and management structure. At the meeting, “Convivio” was adopted as the name for the new network. Under the coordination of Giorgio De Michelis, a proposal was drawn up and submitted to the EC earlier this year.The network currently has 94 members.

40

The Convivio proposal has since received a positive review, although there may be a significant reduction in the funding applied for. Following the reviewers recommendations, the existing workplan is currently being developed further. For more information, contact Giorgio De Michelis ([email protected])

DC The second Jamboree of the EU-funded proactive initiative The Disappearing Computer (www.disappearing-computer.net) will be colocated with the UbiComp 2002 conference (Sept. 29 - Oct. 1, 2002) in Göteborg, Sweden.

January 2001. The exhibition will be open to participants of the UbiComp conference, and will also serve as the basis for the project reviews. UbiComp Conference: http://www.ubicomp.org

The jamboree will include an exhibition showing the results of the 16 DC projects, which all started in

Three i3books in preparation Three i3 books — one on each of the i3 research programmes — are now well underway. Editors-inchief are Patrick Purcell (Connected Community), Dave Snowdon (Inhabited Information Spaces) and John Siraj-Blatchford (Experimental School Environments). All three are currently negotiating contracts with publishing houses.

Each of the books will contain refereed contributions on projects and results in the relevant research programme, as well as articles by guest contributors to give the book more international scope. IIS book webpage: http://www.xrce.xerox.com/communications/i3book/

Community Informatics/Community Networking Research mini-conference The recent 'Shaping the Network Society' Conference, sponsored by CPSR (Computer Professionals for Social Responsibility, formed in 1981- http://www.cpsr.org/) in Seattle, agreed to organize a day-long Mini-Conference on Community Informatics/Community Networking Research on 8 October 2002 in Montreal, Canada in conjunction with the World Forum on Community Networks (Oct. 7 - 12) (http://globalcn2002.org) This is a follow-on to a successful series of research panels organized in conjunction with the Global Congress of Community Networks.

Interested researchers are invited to submit proposals and/or completed papers for peer review for presentation at the conference and for possible future publication as proceedings or as a special issue of an appropriate journal. Appropriate research panels, workshops, and other presentation/discussion formats will be made available to researchers in this field.This conference will also progress the formation of a formalized research CI/CN researchers network.

Media what? MediaX! The Stanford-based Media X programme is inviting proposals for research on interactive technologies. The programme has announced the availability of funding for research on interactive technologies related to sensing and control. Proposals may be for funding of up to $100,000 for one year.

understanding of human psychology/social behavior in the hope of creating new technologies that enable natural interaction with dynamic information and the physical world.

The aim of the initiative is to promote innovative research on the integration of technology and the

www-csli.stanford.edu/MediaX/mediaX.pdf.

For information about Media X see

41

Report

Des visionnaires dans le temple du business1 i3 at Orbit/Comdex Europe 2001 (Basel, 25 — 28 September 2001) Mimo Caenepeel University of Edinburgh [email protected]

“Orbit macht Musik und erzählt Geschichten” “Orbit/Comdex 2001 doesn’t merely astonish with Mega and Giga.The small i3 village in Hall 5 surprises with its collection of projects, which have above all been developed for children and disabled people.” 2 Cradling 25 projects, the i3 Research Village at Orbit/Comdex Europe 2001 did indeed appear “small”, at least in comparison with its broadshouldered commercial/industrial neighbours. The village did not attempt to compete in terms of size, speed and shine. Instead it positioned itself as ‘different’: ‘small’ in the sense of ‘on a human scale’, and unapologetically focusing on human activities such as storytelling, creating and interacting with art, being part of a community, movement and dance, learning, play…. In the context of Orbit/Comdex, that focus was indeed a surprise.

“Ambitions européennes”3

magazine

This was not i3’s first appearance at a major European IT event: some projects already exhibited at the Doors of Perception e-fair in Amsterdam, Educ’mat in Paris, and IST2000 in Nice). But it was definitely the most ambitious attempt so far at highlighting i3 idea(l)s and results in an industrial context that could be overwhelming in places.

Correspondingly visitors, once pulled in, tended to spend a lot of time at the village, often several hours.. “Die Entwicklung wird menschlich”5 As is often the case, the larger impact of an effort like i3’s presence at Orbit/Comdex Europe 2001 is hard to gauge. We can quantify the number of visitors, new contacts and press cuttings. We can reflect on what we learned — about planning and organisation, accommodation, briefing, access, ‘look and feel’, and so on. And all these things matter. But there are also less definable benefits, such as those gained from ‘schmoozing’, as Richard Millwood puts it so nicely,,

“(…) mostly in the evenings, with a variety of notables and interesting people. This informal activity was enjoyable, stimulating and productive, and certainly broader in scope than in the case of previous i3 events.” If organising things together, exhibiting together and feasting together makes us more convivial, than that in itself makes the journey worthwhile. Footnotes 1 Le Matin, 30 September 2001

The presence of a European network at an évenement like Orbit/Comdex Europe was in itself remarkable. The venture definitely broke new ground for i3, and brought with it new contacts for projects, press coverage in different languages, and a steep learning curve for everyone involved. “Les chercheurs tiennent salon”4 54 articles in the European press zoomed in on the i3 Research Village at Orbit/Comdex, the majority of which perceptive and appreciative.Visitors appeared to grasp quickly that what was on show at the i3 Research Village had not been determined in the first instance by commercial considerations and technology push, but had emerged instead from careful observation of the living detail of people’s lives, against the background of a larger vision.

42

2 Basler Zeitung, 28 September 2001 3 L’Express, 25 September 2001 4 La Liberté, 26 September 2001 5 Cash, September 2001

Projects exhibiting at the i3 Research Village Campiello KidsLab Magic Lounge Verbal/non-verbal KidStory Pogo PUPPET Today’s Stories CARESS éTui Playground Ventilar TWI-AYSI/CARE HERE Teleface Unicorn Big brother EYESCUBE Presence/Electronic Graffiti Cyberella HIPS Urban StepStone, ELSNET, i3net, the i3 Summer School and FET were also represented

43

Apart from the Research Village, i3 at Orbit/Comdex Europe 2001 also hosted an i3 conference (with Stephen Heppell and Derrick de Kerckhoven as invited speakers), AGM, project reviews, and a number of social events. Press clippings: http://www.I3net.org/ser_pub/media/pressclip.html Photos: http://media.nis.sdu.dk/image_archive/

magazine

Many people worked tenaciously hard to make the i3 Research Village happen. We would have been nowhere without the organising team (coordinated by Nanett Mosumgaard) and the technological assistance and formidable support of friends like George. But the exhibition was made by all those (i3 researchers in particular) who set it up, were there, engaged in dialogue with visitors, put up with long bus rides at hostile hours, and remained (mostly) equanimous throughout it all. We owe a special thanks to Barbara Strebel and Maria Finders of Orbit/Comdex Europe, who acted as invaluable liaisons, guides and sources of inspiration throughout the process; and, most of all, to Marc Blasband, coordinator of the i3 Partnership Programme, whose generous energy and boundless enthusiasm were matched only by his unperturbable kindness and good humour. Thank you!

44

Column

The ethics of design after September 11 I wrote a first version of the text below in the days immediately following September 11, but the issue of i3magazine it was due to appear in was cancelled. Despite the time that has elapsed since, I have rewritten the original column focusing on the same subject, because I believe that we still need to reflect on how the attacks of September 11 have impacted on us as researchers and designers. I also believe that the interval separating us from the events helps preclude a purely emotional response and enables us to take a longer-term perspective. It was immediately evident to most political observers that the world would not be the same after the terrorist attacks on September 11 2001. Subsequent developments have borne that out. On the one hand, the democratic countries of the Western World have discovered that they were unable to defend themselves against this new type of threat, and that pursuing the terrorists has led them into a new type of war, one with no clear enemies or ending.The resulting anxiety felt by people in the US and Europe may have subsided since, but it is still there, as we could observe in Milano in April when a small plane hit the Pirelli Tower.The need for security and protection is having a deep impact on people’s political orientation and social behaviour. On the other hand, despite the fact that many Muslim countries entered into the coalition against terrorism created by George Bush, many people in Muslim countries (particularly young people, and not just those involved in terrorism) are experiencing growing feelings of hatred towards the US and the Western world in general, and are adopting extremist positions with respect to religion, politics and social life. The Western lifestyle, broadcast all over the world, appears to these people as a provocation, a threat against which they need to protect themselves in order to preserve their traditional values. The resulting unprecedented clash between openness (the chief value of Western democracies, shaping the positive face of globalisation) and diversity (the diverse values most Third World countries want to protect in response to globalisation) seems critical and without solution. As researchers in the area of interactive applications and as designers of people-centred systems, we are (or at least should be) deeply concerned about this situation: systems for supporting remote collaboration and local communities have as their background a world in which people co-exist and

communicate. The attention we have paid (and continue to pay) to the social context of users shows that we cannot design a system that does not reflect the social relations of its users.Yet a world at war does not allow for conceiving of smooth cooperative relations unaffected by political context. So should we begin to pay more attention to security issues?

Giorgio De Michelis University of Milano Bicocca [email protected]

Another relevant issue that affects our research and design work even more is that we always need to take care that our systems support both openness and diversity. This means, on the one hand, that our systems must not enforce any type of separation and exclusion.They must allow access by any type of device and through any type of communication link (independently of its capacity); they must allow for simple ways of extending the number of users; and they must offer different levels of participation to users, depending on the user’s interest and/or involvement in the activities the system supports. On the other hand, our systems must help users to preserve and develop their identity. They should reflect the practices, traditions and values of their community of users; evolve with them; and clearly mark the distinction between members of the community and others. Openness and diversity are difficult to combine: there is an intrinsic contradiction between the former, which requires the abolishment of boundaries, and the latter, which demands a clear demarcation of the community of users. But this contradiction can be resolved if we develop openness and diversity jointly: even while remaining open, a system should support the creation and development of diverse communities; and even while supporting a community and its identity, a system should remain open to newcomers. Some years ago, John Seely Brown and Paul Duguid proposed to support this sort of continuity in systems by providing their boundaries with resources and transforming them into thresholds. Designers, they suggest, should pay attention, beyond the functionalities of their systems, to the resources they put at the systems’ boundaries, to support continuous transition from one mode of use to another, from one user community to another, from open to closed, and so on. In the current situation, this becomes something more than a clever design guideline: it could be the basis for a new perspective on design ethics, and one that we should consider in future work.

45

Book Review

Networked Learning John Siraj-Blatchford University of Cambridge [email protected]

Staples, C. and Jones, C.: .Networked Learning: Perspectives and Issues. Springer, London, 2002. The declared aim of this book is to “(…) develop a perspective on networked learning that tries to articulate the overall relationship between technology and the policies and pedagogies applied in networked learning” (p5). In his Forward, Michael Spector provides an extremely concise and wellinformed introduction to a ‘holistic’ and ‘social’ analytical approach to the study of networked learning.This is a perspective emphasised throughout the book. The book covers a great deal of ground in its 19 chapters and 341 pages, and any special attention given to individual contributions may be misleading. Yet for the sake of brevity some selection must be made. I found some of the papers concerned with studies of networked learning especially valuable, and will focus on these in this review. But I am conscious that in doing so I am neglecting a number of other valuable resources and insights — not least those associated with the identification of communities of practice, institutional readiness in Higher Education, cost assessment, and evaluation.

magazine

Nine of the book’s chapters are devoted to particular studies of networked learning. Vivien Hodgson explores the claims that are made for networked learning in terms of democracy and participation, and finds them questionable. Her analysis of online transcripts suggests that men and women continue to adopt typically gendered ways of talking and interacting and “(…) by so doing were constructing within the conference similar kinds of social identities for themselves as in ‘real’ life” (p240).

Kiran Trehan and Michael Reynolds support this position further in arguing that the rhetoric of empowerment may actually disguise practices that are inherently reactionary. Charles Crooks’ contribution is especially interesting in this respect, as he explores students’ ‘private networked environments’. As many of us have discovered to our cost, students are often ambivalent about any form of orchestrated communication with their peers, and when we introduce networked pedagogies they may resist them. But are networks to be considered as mere delivery systems, or as arenas for community development? Crook cites Sfard (1998) and offers the credible prospect of “(…) learning as ‘participation’; such that education is viewed as an inherently community-orientated activity” (p306). The text ends with Christine Steeples and Chris Jones discussing the future potential for networked learning. This final chapter offers readers an extremely helpful analytical framework for networked learning. In their ‘Pedagogical framework’, the authors distinguish between four elements: ‘Philosophy’, a ‘High Level Pedagogy’, a ‘Pedagogic Strategy’ and ‘Pedagogic Tactics’. In many institutional contexts the philosophy of learning remains implicit; yet underlying beliefs about learning need to be articulated to inform the design of the networked environment, and will inevitably determine the kinds of activities afforded by it. High-level pedagogy is distinguished from strategies and tactics to highlight the crucial difference between the application of abstract pedagogic conceptions such as ‘problem based learning’, broadly defined plans, and the specific actions through which these plans are effected. Drawing upon work by Goodyear and the NlinHE team (2001), Steeples, Jones and Goodyear also present a model for organisational contexts that relates these Pedagogical frameworks systematically to the real-world networked learning realities that are defined on a day-to-day basis by the specific concrete tasks and technologies applied by students. This is a model that may be usefully applied to a variety of research and development contexts. It also has considerable merit in offering a reflexive framework for the joint development of research and practice. References Sfard, A. (1998) On two metaphors for learning and the dangers of choosing just one, Educational Researcher, 27, pp4-13. Goodyear, P. and the NLinHE team (2001) Effective networked learning in higher education: notes and guidelines. CSALT, Lancaster University (http://csalt.lancs.ac.uk/jisc/advice.htm).

46

Future events Links to all events on this list are available at http://media.nis.sdu.dk/conferences/ European Workshop on Mobile and Contextual Learning 20-21 June, 2002 University of Birmingham, UK ED-MEDIA 2001 - World Conference on Educational Multimedia, Hypermedia & Telecommunications 24-29 June, 2002 Denver, Colorado, USA International CLASS Workshop on Natural, Intelligent and Effective Interaction in Multimodal Dialogue Systems 28-29 June, 2002 Copenhagen, Denmark Assets 2002 - The Fifth International ACM SIGCAPH Conference on Assistive Technologies 8-10 July, 2002 Edinburgh, Scotland First International Symposium on Collaborative Information Visualization Environments 10-12 July, 2002 London, UK KMN02 - IEEE International Workshop on Knowledge Media Networking 10-12 July, 2002 Kyoto, Japan SIGCHI-NZ Symposium on Computer-Human Interaction 11-12 July, 2002 University of Waikato, Hamilton, NZ Embodied conversational agents let’s specify and evaluate them! 16 July, 2002 Bologna, Italy The Third Wireless World Conference: The Social Shaping of Mobile Futures 17-18 July, 2002 University of Surrey, UK ECAI 2002 - 15th European Conference on Artificial Intelligence 21-26 July, 2002 Lyon, France Summer School on Ubiquitous and Pervasive Computing 7-14 August, 2002 Schloss Dagstuhl, Germany ICCC 2002 - 15th International Conference on Computer Communication 11-14 August, 2002 Bandra, Mumbai, India PRICAI-02 - The Seventh Pacific Rim International Conference on Artificial Intelligence 18-22 August, 2002 Tokyo, Japan

WMTE 2002 - IEEE Intl Workshop on Wireless and Mobile Technologies in Education 29-30 August, 2002 Växjö, Sweden COSIGN 2002 - 2nd Conference on Computational Semiotics for Games and New Media 2-4 September, 2002 Augsburg, Germany ICALT 2002 - IEEE International Conference on Advanced Learning Technologies 9-12 September, 2002 Kazan, Russia Fourth International Symposium on Human Computer Interaction with Mobile Devices 18-20 September, 2002 Pisa, Italy ICDVRAT 2002 - Fourth International Conference on Disability,Virtual Reality & Associated Technologies 19-21 September 2002 Veszprem, Hungary UbiComp 2002 - The Fourth International Conference on Ubiquitous Computing 29 September - 1 October, 2002 Göteborg, Sweden ACM Collaborative Virtual Environments 2002 30 September - 2 October, 2002 Bonn, Germany 1st International Workshop on 3D Virtual Heritage 2-3 October, 2002 Geneva, Switzerland IEEE 4th International Conference on Multimodal Interfaces 14-16 October, 2002 Pittsburgh, PA, USA 7th ERCIM Workshop on User Interfaces for All 23-25 October, 2002 Paris (Chantilly), France APCHI 2002 - 5th Asia Pacific Conference on Computer Human Interaction 1-4 November, 2002 Institute of Software, Beijing, China IADIS International Conference WWW/Internet 2002 13-15 November, 2002 Lisbon, Portugal ICCE 2002 - International Conference on Computers in Education 3-6 December, 2002 Auckland, New Zealand

47

i3net member sites Austria University of Vienna Belgium European Schoolnet Office Linc vzw Public Library of Turnhout Starlab (Closed 12-06-2001) Universitè de Liëge Vrije Universiteit Brussel Denmark Aalborg University Aarhus University LEGO System A/S Soundscapes Studios The Danish Isles - User Community UNI-C University of Southern Denmark, Main Campus: Odense Finland Åbo Akademi University Helsinki University of Technology Helsinki University of Technology Nokia Research Center France Cryo-Interactive ENST Bretagne LIMSI-CNRS Université Sorbonne Paris V Xerox Research Centre Europe Germany Bremen University Competence Center Softwaretechnik Fraunhofer IAO Fraunhofer Gerhard-Mercator-Universität German Research Center for Artificial Intelligence (DFKI) GMD - Forschungszentrum Informationstechnik GmbH Media World GmbH & Co KG Ravensburger Interactive Media Transfer Center Global Working at DFKI Universität Dortmund ZKM Zentrum für Kunst und Medientechnologie Greece Computer Technology Institute Computer Technology Institute FORTHnet ICS-FORTH Lambrakis Research Foundation Municipality of Chania Technical University of Crete University of Athens, School of Philosophy University of Patras University of the Aegean Ireland University College Dublin (UCD) University of Limerick Israel Ben-Gurion University of Negev

48

Italy Alcatel Italian Comune di Reggio EmiliaAssessorato all Educazione e Consiglio Nazionale delle Ricerche Consorzio Roma Ricerche DISCO - University of Milano Bicocca Domus Academy srl Innovative Devices & Engineering for Automation (IDEA) ITC-IRST, Istituto per la Ricerca Scientifica e Tecnologica Scuola Superiore S Anna (SSSA) Siena University (DII) Siena University (MCL) SKYDATA Universita degli Studi di Bari University of Milano

Norway Human Factors Solutions (HFS) SINTEF Telecom and Informatics Telenor R&D Portugal CNOTINFOR - Centro de Novas Tecnologias da Informação Instituto de Engenharia Slovak Republic Comenius University Spain IIIA - CSIC Consejo Superior Investigaciones Cientificas REM Infographica Universidad Politecnica de Madrid University of Barcelona University Pompeu Fabra Sweden Alcesys AB Enlight AB Göteborg University Halmstad University Högskolan för läraritbildning & kommunikation Högskolan i Halmstad Landskrona Kommun Royal Institute of Technology (KTH) Swedish Institute of Computer Science (SICS) Uppsala University Switzerland EPFL - Computer Graphics Lab University of Geneva The Netherlands Compuleer Doors Of Perception Halmstad University LOST BOYS content & usability Meru Research Netherlands Design Institute (Closed 31-12-2000) Philips Design Philips International Technische Universiteit Eindhoven University of Twente UK 3D Scanners Ltd Apple Computer Benelux British Telecom Laboratories Crystal Presentations Glasgow University Homerton College IDEO Product Development Illuminations Television Imperial College of Science and Technology Lancaster University Logotron Napier University Queen Margaret University College Royal College of Art Sussex University Ultralab University of Bristol University of Edinburgh University of Edinburgh University of Leeds University of London University of Manchester University of Nottingham University of Nottingham University of Reading University of Strathclyde University of Sunderland

i3net The European Network for Intelligent Information Interfaces i3net, the European network for intelligent information interfaces, was created in 1996 to explore visionary, humancentred interactive systems for people in their everyday activities. The global vision pursued by the i3 community is to help invent and shape the future through comprehending the basic structures and trends in society and in the life of the individual. Based on that understanding, i3 focuses on research lines that investigate new relationships between technology, people and design. i3net has supported three research programmes: Connected Community (1997-2000), Inhabited Information Spaces (1997-2000) and Experimental School Environments (1998-2001). More recently it has also begun to adopt relevant organisations outside those programmes as member sites, and built links with closely related EC initiatives such as the Disappearing Computer (2000 – 2003). All this brings the current size of the community to about 450 researchers from around 150 organisations, one third of which are companies. The mission of the current i3 interim network (March 2002-February 2003) is to provide continuity for the community while a proposal for a new network, supported by a broad panel of i3 members, is developed and negotiated.The name of this new network will be Convivio.

i3net Coordinating Group (CG) Riccardo Antonini (Italy) Niels Ole Bernsen, (coordinator, Denmark) Tony Brooks (Sweden) Mimo Caenepeel (UK) Erik Granum (Denmark) Ulrich Hoppe (Germany) Jean-Claude Martin (France) Giorgio de Michelis (Italy) Richard Millwood (UK) Alan Munro (UK) Patrick Purcell (UK) Thomas Rist (Germany) Norbert Streitz (Germany) Jakub Wejchert, observer for the Commission Observers for the Disappearing Computer projects: Lorna Goulden (Netherlands) Spyros Lalis (Greece) Paddy Nixon (UK)

Natural Interactive Systems Laboratory University of Southern Denmark Main Campus: Odense University Science Park 10 5230 Odense M Denmark Tel: (+45) 65 50 35 44 Fax: (+45) 63 15 72 24 Email: [email protected] URL: http://www.i3net.org/ i3LabTV: http://www.i3net.org/i3labtv/ i3net Secretariat

Niels Ole Bernsen co-ordinator Mimo Caenepeel i3magazine Svend Kiilerich, i3net manager Merete Bertelsen secretary

magazine

is published at the Human Communication Research Centre, University of Edinburgh, Scotland using QuarkXPress™ Design by the Domus Academy, Milano Layout by Junction, Edinburgh Editor Mimo Caenepeel ISSN: 1397-906X © i3magazine 2002 Contributions to i3magazine should be sent to [email protected] Tel: +44 131 650 4594 Fax +44 131 650 6626 Address corrections should be sent to [email protected] Fax: +45 63 15 72 24 Print run of this issue: 1000 Material for the next issue is due: 30 August 2002

http://www.i3net.org