hypertexte_paris

Amanda Wright : C'est une perle de Londres. Formée à la littérature et à la philosophie à Cambridge et Birmingham ainsi qu'à l'informatique à Warwick, cette jeune scenariste est aussi habitée par un rêve partagé par le Groupe Empirical Modelling de Warwick (UK) qui est dirigé par Meurig Beynon et Steve Russ. Amanda s'envole pour la Chine le 13 février 2006 et laisse derrière elle sans regrés sa collection de centaines de pandas en peluche.

Delphine Richer : Etudiante de 4ème année à L'Ecole Nationale des Beaux Arts de Bourges, Delphine s'interesse à la question de la (non) propriété de l'oeuvre à l'artiste. Ses dispositifs engendrent un rapport entre le public et un objet co-créé par ce public. Delphine est fachée avec la beauté et les apparences en Occident. Elle partira en Chine en avril 2006 pour condidérer le sort de l'esthétique au coeur de l'Empire du Milieu, notamment dans les arts martiaux.

Greg : Artiste issu de la Villa Arson (ENSA de Nice), GregBot se distingue par son goût pour le décloisonement des pratiques artistiques contemporaines vers les mondes de la science, du design, de l'entreprise et de l'économie en général. Partagé entre le détournement de l'outil économique à des fins artistiques et l'option plus difficile de l'invention de nouveau outils pour la création, GregBot s'envolera au mois de Mars 2006 pour la Chine pour y explorer de nouveaux rapports à la contrainte.

EM créé des modèles. Ces modèles sont constitués de parties indépendantes que l'on peut enlever ou rajouter sans détruire le programme formé. EM permet de qualifier le programme de vivant par opposition aux logiciels courants qui ne peuvent pas être évolutifs. Modèle Erlköning : L'intérêt de ce modèle pour le lien avec la création artistique est qu'il se base à la fois : 1) sur un texte (le poème de Goethe), sur une musique et sa partition, sur le temps, et sur les couleurs qui permettent de visualiser le procédé du modèle.

Modèle OXO : Ce modèle à l'avantage d'être très simple pour expliquer les principes d'EM. Il se base suivant le jeu du Morpion. http://www.dcs.warwick.ac.uk/modelling

Modèle des Trains : L'aspect narratif de ce modèle peut faire office de lien avec le projet P3 de Mètis. Stimulés par la présentation philosophique et technique de l'EM, les participants seront invités à contribuer à la création de leurs outils et du film advertainment P3 (projet pilote de Metis) dont les processus et résultats seront présentés à Pretoria, Beijing et Paris entre Mai et juillet 2006 dans cadre de forum internationaux scientifiques, technologiques et artistiques. Litterature, Art-Cinéma-Hypermedia, R&D, Science et Ethique seront au cour du débat. Mètis est un réseau global constituant une équipe virtuelle interdisciplinaire -Arts, Sciences,Technologies, etc- visant à explorer la créativité, l'expression et l'innovation collective. Mètis, coordonné par l'oper'acteur dans l'art Rody Klein, est constitué de volontaires spontanés pour créer un "film" advertainment " de type advocacy contribuant à faire des Jeux Olympiques de Beijing 2008, un évènement qui favorise un rapport plus "durable" entre les besoins d'expression au coeur de l'Esthétique et les impératifs de la production industrielle, souvent perçus comme irréconciliables. http://www.arenotech.org

Metis meets Empirical Modelling: from ancient wisdom to emerging technology
Rody R. KLEIN, SysCom Lab, Université de Savoie, Le Bourget du Lac, 73010, France
^{Meurig BEYNON,}Computer Science, The University of Warwick, Address, City, Postcode, UK

Abstract: The Metis project brings together a global virtual team for the collective creation of an advocacy advertainment film for the 2008 Beijing Olympics. This team includes computer and human scientists, artists, NGO officers, students, global Olympics volunteers and industrial partners from Europe, Africa, China and North America. The ambitious scope of Metis demands a collaborative technology that can address the intercultural mediation of creative and innovative processes and that is highly adaptive to unstable contexts and unpredictable situations. As is suggested by its motto: ‘Shooting the film while building the camera’, Metis envisages the development of such a collaborative technology and infrastructure as part of the concurrent responsibility of the global virtual team. This paper reviews the progress of the Metis project to date, with specific reference to the broad challenges being faced in deploying current technologies for computer-supported collaborative working and learning (CSCW/CSCL). These challenges, which stem from the diversity of the cultures, contexts and processes for creation, communication and production represented in Metis, relate yet more broadly and generally to topical issues concerning eAdoption worldwide. The paper briefly discusses the prospects for addressing these issues with reference to the future emerging technology of Empirical Modelling.

Keywords: Digital divide, NGOs, Innovation, CSCW, CSCL, Interface Design, Empirical Modelling (EM), Humanities computing, Creativity, Developing countries, Capacity Building, Cultural Context, Media content, Olympics.

The Metis project [32] is a wide-ranging exploratory process that aspires to escape the usual logistics of participating laboratories. Empirical Modelling (EM) is an exploratory project [30] directed at developing an alternative view of computing that is likewise wide-ranging. This paper outlines how broader views of technology and creativity with cultural roots in ancient Greece and China that are topical for Metis can be supported by EM principles and tools. This potentially has wide implications for cross-cultural projects that involve computer-supported collaboration across the digital divide.

The orientation of the Metis project is profoundly influenced by ethical concerns. A key objective for Metis is the promotion of the broader vision for the Olympics that is embraced by the Humanistic Olympic Study Center [31]. The vehicle for this promotion is an advocacy movie in the form of advertainment: a blend of advertising, artistry and entertainment. Such a movie will advocate a conception of the Olympics where benevolent motivations for global good conjoin with passion for sports, arts and dedication to one’s country – an antidote to the merry commercial and professional process to which the Olympic dream is in danger of being reduced.

Addressing the ethical agenda of Metis has far-reaching implications. Setting up the Metis project involves bridging gaps between the perspectives of science, art, commerce and technology. Its Olympic vision embraces global culture, peace and respect for the environment, but must also emphasise the importance of protecting the rich diversity amongst individuals, communities and ecologies. These vital messages are well perceived in the world: they constitute a powerful reserve of hope and humanistic expression which could be skilfully mediated by new information technologies. But our technologies cannot be based solely on the dominant culture and ideologies of the Western world.

The elaboration of the concept behind Metis can be seen as drawing on the two cultures that come together explicitly in the Beijing Olympics: the Chinese and the Greek. The importance of these cultural contributions is that they give insight into respects in which our accepted understanding of key concepts – such as creativity and technology – has become limited, perhaps in somewhat the same way that the modern conception of the Olympics has diverged from that of the ancient Greeks. These limitations motivate the practical approach adopted in the project, to be described in section 3.

The name Metis itself derives from the Greek word for 'wisdom'. In their analysis of the concept, as it relates to ancient Greek culture, Détienne and Vernant [12] observe that: There is no doubt that mêtis is a type of intelligence and of thought, a way of knowing; it implies a complex but very coherent body of mental attitudes and intellectual behaviour which combine flair, wisdom, forethought, subtlety of mind, deception, resourcefulness, vigilance, opportunism, various skills, and experience acquired over the years. It is applied to situations which are transient, shifting, disconcerting and ambiguous, situations which do not lend themselves to precise measurement, exact calculation or rigorous logic. Détienne and Vernant contrast this broad concept with the Modern Greek scholars' image of Greek thought – an image "in which mêtis is conspicuous by its absence". This contrast underlies the crucial difference between the path of the ancient Greek Technite and that of a typical contemporary engineer – the modern technician who has been disconnected from a cosmogonist view of the world because of their extreme specialization.

In the Metis project, the authentic concept of mêtis guides the development of a collaborative technology and shapes its associated interactive environments. Of equal importance is the conception of the collaborative activity itself, and the nature of its products. The notions of 'creativity' and 'innovation' have particular relevance (see [16]), both because of the artistic aspect of advertainment and because of the need to accommodate a wide variety of cultural perceptions of these terms.

According to Lubart [23]: Creativity is the capacity to create something which is both new (original, unexpected) and adapted (useful, adapted to constraints). This pragmatic North American definition may be contrasted with the European conception of "being creative", which is understood metaphorically as referring to the willingness to express oneself freely without any practical aim in mind, almost detached from reality [14]. In so far as these two definitions can be construed as referring to a common activity, there are complex socio-political reasons why what is a scientific concept in USA is a metaphor in Europe. As Joas [14] has observed, such common construal is possible provided that the phenomena that are implicitly invoked in 'being creative' can be the target of a real experience.

As described in more detail in [16], the work of the sinologist-philosopher François Jullien [15] is helpful in looking at Western innovation and creativity from a Chinese point of view. Debray [11] and Latour [20] further stimulate the orientation towards China. Where the Western traditions of thought attach great importance to originality and the contribution of self, Tao recognises only the harmony of individuals on the path originating in the consciousness of belonging to nature. This constitutes the fundamental recognition of correlativity: continuity and connexion between every single thing. In such a philosophical context, innovation is conceived differently – as making the relationships between things more readily experiencable by presenting them in such a way that they are more easily accommodated in the mind of the viewer. In contrast, the Western conception of innovation involves performance that purports to create novelty, and – by implication – promotes a vision of self and present reality that violates the Tao idea of all reality as infinite [15].

The above account of the context for the Metis project has been amplified in several previous publications [17, 27]. It highlights commonplace pitfalls in cross-cultural projects involving information technology. In relation to the Beijing Olympics, to follow the established practice in promoting the Olympics could be viewed as imposing alien cultural values where both the vision for the Olympics and the perception of innovation and creativity are concerned. At the same time, developing an alternative practice that does fuller justice to a broader Olympic vision is challenging on account of the very practices and technologies that have been developed to sustain Western cultural traditions. This has been the motivation for incorporating 'building the camera' with 'shooting the film', a notion whose consequences are the subject of the next section.

· recruiting an intercultural global virtual team of volunteers composed of researchers in computer science, human sciences, movie professionals, artists, industrialists and students inclined towards interdisciplinary research.

· carrying out an advertainment cinematographic project within the framework of Beijing Humanistic Olympic Games 2008.

· developing – through an ongoing iterative process of conception and test – a collaborative platform for the creation, follow up and mediation of the Metis project’s artistic, scientific, technological and commercial components.

To date, the Metis project has explored several complementary aspects of its broad agenda, relating to issues such as Innovation Management [8], Ergonomy [25], Multimedia Document Management [8, 9], Education [19] and the technological needs of NGOs [18, 19]. This research has been aimed at clarifying how to organize such an innovative approach, both from the artistic, technological, educational, commercial and research point of view. Some comparative study of aspects of collective creativity in China and other cultures has also been carried out [27].

The guiding principle behind the Metis project is that – in order for contemporary actors to exercise their artistic creativity and exploit radical technological innovation to the full – their interaction should be mediated by an appropriate interface through which they are sensitized continuously to what makes the core of humanities: mêtis. This entails enhancing awareness beyond the scope so far envisaged in the CSCW literature – in ways that are not merely concerned with enriching the interaction between the members of the virtual team, but also with increasing each participant's reflective perception and self-awareness during the creative process. The intended effect of this enhancement is twofold. It broadens the focus of attention so as to take account of a larger and more comprehensive set of variables that are deemed to impact on the ongoing creative process. It also augments the role of the computer so that, in addition to supporting the automation and repetition of routine tasks, it can help to address the uncertainty which is highly specific to innovation and creative processes.

The proposed tool for enhancing awareness is a Trace Composer. The function of this tool is to support information capture and integration in order to help the team members during the creative process. To this end, the Trace Composer will increase feedback that relates both to individual and collective activity. The monitoring of each virtual team member’s activity is to be both voluntary and assisted. Using the metaphor of the "mirror", the team member should be able to express – as far as possible – information about their own cognitive, emotional, and conative state. This kind of information is intended to improve awareness of meta-cognitions and hidden cognitive automatisms. The objective is to plunge the team member into a holistic view of their activity "in context", giving the kind of open systemic view that activity of an analytical nature rarely provides. In relation to teamwork, it enables each team member to have some idea of the state of mind and direction of thinking of other members. This can help to ensure that a new idea is not simply generated 'spontaneously', but is connected to the context and adapted to current constraints. It can also raise the profile of inappropriate hidden assumptions and automatic responses that can be constructively challenged. This potentially stimulates a re-generation of the creative process that leads to greater freedom of choice or flexibility, and a better understanding of the relationship between original ideas and valid constraints.

So complex and mysterious is the creative process that capturing every associated influence and aspect, both individual and collective, is out of the question. Yet, as Fischer [13] observes: "The power of the unaided individual mind is highly overrated. Much human creativity is social, arising from activities that take place in a social context in which interaction with other people and the artifacts that embody collective knowledge are essential contributors.". Metis aspires to a CSCW environment that can help to identify the exact nature of this contribution, and so enhance it.

Where the influence of group interaction on individual creativity is concerned, there is an important distinction between synchronous and asynchronous activity. Some influences are present as immediate perceptions of feelings and ideas and relate to current "synchronous" activity. Other influences that are perceived in the present relate to "asynchronous" activity, performed in the past, but having direct relevance to the current context. Asynchronous activities that provide access to useful information during the immediate creative process are typically associated with lengthy experimental procedures. Such information may be potentially misleading, since it was collected at another time and within a different context. It may not necessarily take a handy form that can be exploited in fast short-term creative processes. Traces of such activities may still contribute to the generation of archives to be used at a later stage.

Developing the Trace Composer presents major technical challenges. In order to promote cross-cultural communication and maximise the immediate impact of synchronous and asynchronous activities, it will be necessary to create a suitable medium. To this end, quantitative and qualitative traces of interactions (via the keyboard, mouse, screen and peripherals) should be transformed from typically dense textual information into visual information. The use of cues such as colour, link, thickness and shape facilitates understanding much more effectively than do texts, patterns, rules and numbers. But while visualization techniques, quantitative observations, and traces of the activity involving team members, the server and the clients are important, it is no easy matter to extract meaningful information from them. Appropriate instrumentation is needed to mediate the traces of each team member's local activity, and raw logs of the server activity. Communication between one member and another during a collective or individual creative session poses a particular difficulty. Such interaction is not usually observed on the server, but may be an important ingredient in explaining the member’s direction of development and train of thought. It then seems that, rather than instrumenting a team member’s computer so that their interaction is observed, it is preferable to identify diverse roles (observer, self-monitor, group-monitor, learner, mediator) and ask each member to provide explicit information about their current mode of activity.

The technology that is ideally required to support this vision for communication is quite unlike the document-based interaction that is characteristic of many CSCW environments. Signals about team members' activities should be constantly sent in a form that does not require a transition to phonetics, but establishes augmented consciousness through exploiting dynamic figures, pictograms or icons. If such an idea is hard for orthodox alphabet users to conceive, it is far more plausible to Chinese users, who are accustomed to learning to read new sinograms very easily. For instance, WuBi – recently introduced in Chinese – is a precedent for purely visual communication of this nature, involving translation from sign to meaning without any transition through phonetics. Such communication could be supported by developing a specific device, similar to the WuBi Chinese keyboard that allows the direct entry of Sinograms(see Figure 1).

An essential feature of the communication medium that underlies Metis is its capacity to evolve as the project develops. It is envisaged that all the members of the global virtual team will participate actively in the development and experimental use of the Trace Composer. Conceptual support for such a vision can be found in the research of Pierre Levy [21], who seeks to capture the way in which experience is conditioned by the frames and forms of the surrounding culture that shapes the ever-evolving nature of meaning in context. Levy's Information Economy Meta-Language (IEML) [28] – a product of his research on "dynamic ideography" – may provide the theoretical background for the challenging task of integrating the interdisciplinary information. The traces of the Trace Composer, multimedia in nature, might then comprise a set of linked units, each composed of an expression of subjective intention, a fixed or dynamic image, sounds, music, a conceptual referential from IEML, and an optional compendium of items (such as an object, drawing or storyboard) that increases the chance of conveying the meaning of "any complex personal experience".

Empirical Modelling (EM) is an approach to computer-based model-building that is based on three primary concepts: the observable, the dependency and the agent [30]. EM is seen in the first instance as a means of creating interactive artefacts and environments that serve a role in shaping the modeller's personal understanding of a situation or phenomenon. The core activity in EM is the construction of families of definitions ("definitive scripts") that resemble the abstract network of definitions that relate the cells of a spreadsheet. Each definition attaches a formula to a variable. This formula may specify the current value of the variable explicitly, as in a conventional programming assignment, or give a recipe for computing a value from that of other variables. The effect of redefining a variable (cf. redefining a cell in a spreadsheet) is then to 'instantaneously' update the values of all variables whose values are directly or indirectly defined with reference to it.

The modeller's interaction with a definitive script is the archetype for all interpretative processes that are associated with EM. Each variable corresponds to some entity external to the computer model that is an observable, in the sense that it has a distinguishable identity and a current value or status: the shadow of an object, the age of a person or the openability of a door. Each definition corresponds to an association between one change to the value of an observable and that of another that has – or can be deemed to have – the qualities of a mechanical linkage. Relationships between observables of this nature are pervasive in experience, and are called dependencies. Simple examples include the relationship between the position of a light source, an object and its shadow; the relationship between current year, date of birth and age; the relationship between whether a door is locked and whether it can be opened. In modifying a script, the modeller will add or remove variable definitions, or redefine an existing variable by substituting a new formula for it. Such an action on the part of the modeller corresponds to exercising some form of agency that is in general to be interpreted with reference to the external situation. For instance (referring to the examples given above), the modeller relocates the light source, updates the current year, or redefines the door that is now jammed as unopenable irrespective of whether it is unlocked.

EM puts modelling-with-definitive-scripts to a most general purpose: that of mediating understanding of any situation from the perspective of any agent. In this context, the term 'agent' is used to refer to anything that might be responsible for changing the current values of observables in a situation, whether this is a person, a device (such as a door spring) or a natural agency (such as gravity or the wind). The primitive interaction of modeller with script is in itself only sufficient in situations where relationships between observables and changes to their values can be studied in the absence of other agents, as in controlled experimental situations. Nonetheless, the same principle that is used to predict the shadow cast by a ray of light falling on a needle in the optical laboratory still applies to modelling the shadow of a sailing ship's mast in the light of a moon. To make the generalisation, it is 'only' necessary to model the ways in which the orientation of the ship and the locations of clouds, sails and sailors interact with the underlying dependency between illumination and shadow.

A full elaboration of EM principles and tools is beyond the scope of this paper (see [30]). A simple illustration, drawn from previous work specifically directed at using EM in CSCW, will be used to outline how EM might be applied to the practical Metis agenda.

Figure 2 is a screenshot from an EM artefact (see railwayYung1995 in the EM archive [29]) that was developed for tutorial purposes in connection with various aspects of railway modelling. Three different modelling perspectives are integrated within the model. A simple model railway layout, together with a control interface such as a model railway operator might use, features on the left of the figure. The depiction of the actual trains controlled from this interface is inconsistent with the model railway interpretation; it reflects instead the perspective of an observer in an actual electronic signal box, where the track segments that are currently occupied by trains are highlighted. On the right-hand side of the figure, animated stick figures depict the participants in a train arrival and departure protocol, showing the interaction between the actions of stationmaster, the guard and the passengers. Note that the toy railway and the protocol animation were first developed independently – the schedule for visiting stations that features in the animation (train 1 travels from station A10 to B5 to A5 cyclically) was instantiated by identifying these stations with particular track pieces.

The current state of the hybrid model shown in Figure 2 is determined by an extensive definitive script, comprising several hundred observables. Examples of observables include the location of the track pieces, the direction to which the track points are currently switched, the positions of trains 1 and 2, the positions of the buttons in the screen interface, the locations of the stations A5, A10 and B5, the status of the doors on train 1, locations and destinations of passengers, the abstract point currently reached in the protocol of interaction between the stationmaster and the guard. Dependency also features in many aspects of the model: the perceived location of a passenger on the train is determined by the location of the train; the status of the lines depicting the orientation of the points on the display is determined by the setting of the points; the status of the railway operation depends on whether trains have crashed into each other or into the track points.

All manner of agency, both explicit and latent, is represented in the railwayYung1995 model. As explained above, the role of the model-builder is perhaps most appropriately conceived as integrating that of several designers acting in different roles. The redefinitions that are accessible via the interface at the bottom left determine agency in the traditional computer user mode, and approximate that of a model railway operator. A more realistic representation of this role would allow the operator to "pick up" a train and replace it elsewhere on or off the track: definitions to accomplish this directly can be entered into the model through the input window of the interpreter. The roles of agents that are internal to the animation (the trains, passengers and railway staff) are also mediated by definition, but in these cases redefinition is automated so that the movement of trains and the actors within the model can be simulated as if it were being carried out manually but without requiring any interaction on the part of the modeller. For this purpose, the actions of passengers and railway staff, though faithful to their intentions and protocols, are scheduled with an appropriate degree of randomness in respect of specific time of acting. In contrast, trains move from one track segment to the next in accordance with the abstract connectivity pattern specified by the combinatorial diagram displayed within the track layout on the left of Figure 2, in a fashion that is strictly clocked.

The railwayYung1995 model illustrates in microcosm the potential for cross-cultural modelling of interactions between CSCW participants. Variants of the interpreter deployed in Figure 2 allow the roles of different agents to be distributed. This could be exploited to allow the roles of passengers and railway staff to be played by human agents, for instance, or to enable the various strands of modelling activity in the hybrid model to be separated and developed independently subject to looser constraints on collaboration.

Previous research on EM has examined the scope for deploying EM principles and tools in concurrent design from a technical perspective (for details, see [2,3,4]). Subsequent development has focused on developing a better conceptual foundation for EM principles, with particular reference to the role that EM plays in integrating the roles of human and automated agency. This section identifies key respects in which the orientation of EM resonates with the Metis project and briefly discusses how EM relates to the practical aspirations of Metis, as outlined in section 3.

The first obstacle to a clear understanding of EM is semantic in nature. Reflecting on the significance of an EM model leads to the conclusion that it is inappropriate to regard the meaning of such a model as being specified by a set of functional behaviours – as is the meaning of a computer program. By way of illustration, the EM railway model depicted in Figure 2 is to be identified with the definitive script that specifies its current state, as captured in the figure and by the values of all the values and formula attached to the observables as of that moment. It may be tempting to infer that this script can be read as a set of propositions, but this mode of interpretation is fundamentally unconvincing. For instance, perhaps train 1 is now at B5, but this is no guarantee that it will be always at B5. The train may move, or be picked up and replaced elsewhere on the track, or be redefined so as to become a ghost train such that other trains can pass through it without crashing. By citing each of these scenarios, all that is meant is that the 'very same script' could, by dint of minor redefinition of observables such as could be performed by agents in different roles, be transformed into a new state that it would be unreasonable to deem to be 'a new model'. Uncomfortable as this semantic notion may seem at first sight, it is the only viewpoint that can accommodate the ill-defined concept of 'meaning in the making' and 'meaning in the process of being negotiatied' that has to account for how the railway model evolved to its current state, and how it even now remains so obscure in relation to possible interpretations. It is impossible to say decisively whether B5 is a track piece in a toy railway layout, or a track segment in an improbably contrived actual railway. Either interpretation can be made more or less plausible according to what future trajectory of agent actions is played out whether manually or automatically.

The nub of the semantic difficulty in this context is a philosophical problem. By directing redefinitions in the script of Figure 2; by removing or introducing additional observables, dependency and agency; by exercising discretionary privileges to suspend any automatic activity; by invoking whatever imaginative interpretations the current state of the model can legitimately bear; the modeller can make of the railway model not merely a model of a toy railway, actual railway, or train arrival and departure protocol, but a model of whatever takes their fancy. For the legitimacy of the modelling exercise, all that is required is that the continuity and connexion between the experience of the EM railway model as a railway and the experience of the 'self-same' model as something else altogether should be recognisable – to the modeller at any rate. And though it is natural to challenge the subjective nature of this characterisation, it is difficult to imagine any other evidence for its authenticity apart from the ability of the modeller to trace and retrace the path of interaction and interpretation followed in transforming the model from one place to another, and to account for this in a coherent and consistent manner. In this process of EM, the connection with creative activity in the Tao sense to which section 2 alludes is immediately apparent.

The openness of the modeller's interaction and interpretation of an EM model is not only significant in relation to the attribution and evolution of meaning. It is the basis on which an EM model can serve as a practical construal (cf. [30, 7]), embodying the patterns of observables, dependencies and agencies that the modeller believes to be valid in the situation to which it refers. In so far as an EM model is a good construal, it will prove effective in accounting for all manner of patterns of interaction encountered in a situation, even though these may not have been explicit in the initial conception. By way of illustration, it would be a relatively modest task for the modeller to introduce suitable definitions into the railway model on-the-fly to reflect the possibility that whilst train 1 was in motion from station A10 to station B5 it became necessary (as a result of some emergency) to stop the train for passengers to disembark instead at B1. The destination of a passenger could also as of that moment – as if in response to a general announcement – be redefined to that station location known to be nearest to the passenger's home, so that conceivably a passenger originally intending to travel to B5 would instead travel on to A5. This hints at the sense in which EM accommodates that broader notion of intelligence and thought that can be applied to "transient, shifting, disconcerting and ambiguous situations [that] do not lend themselves to precise measurement, exact calculation or rigorous logic" [12].

Previous EM research strongly suggests that modelling with construals has highly significant advantages over any variety of computer programming where activities that involve an intimate integration of manual and automated activity are concerned. Whereas the interpretative framework that surrounds a conventional program is fixed on conception, the whole purpose of the construal is to reflect relationships that are fluid and negotiable. In contrast to programming, where there is a sharp distinction between development and use, it is a matter of interpretation whether a redefinition is associated with 'shooting the film' or 'building the camera' (cf. [1,24]). Research on themes relating to concurrent engineering [2,3,4], decision-support [5], educational technology [6,7], humanities computing [24] and on an underlying philosophical framework for EM [1] all points to the conclusion that modelling with construals supplies an essential bridge between pre-articulate human experience and interaction based on language. In particular, in a collaborative environment, direct interaction with another's construals (cf. [10]) is a form of pre-linguistic communication that arguably has a fundamental role to play in boot-strapping the more radical high-level forms of linguistic interaction envisaged in the Metis project.

Though there has been limited external adoption of EM principles, and practical experience of EM has been largely confined to university project work over a fifteen year period, there is a considerable body of informal evidence to indicate that EM can help in addressing technical issues, topical for Metis, that have proved problematic in relation to existing technologies. A pervasive problem – that of adapting software to different platforms [26] – was resolved in the specific context of cross-platform broadcasting by work at the BBC R&D laboratories in which dependency-based models of content featured prominently. The readiness with which extant models built by one modeller have been revisited and extended by another – sometimes in a casual off-the-cuff manner, illustrates the potential for models of cooperative development that are better suited to Metis's vision for a global virtual team, where only a modest investment can be expected from each individual. Finally, the vigorous accumulation of models, refinement of tools and enhancement of skills that has been manifest in EM project work, despite the modest resources that have been available to support this activity, suggests as yet unrealised potential for capacity building that is based on informal cross-cultural interaction.

EM supports the broader conception of creativity identified by Metis as crucial to deeper cross-cultural East-West collaboration. By focusing on modelling with construals, there are excellent prospects for developing continuously evolving CSCW environments that allow modes of communication based on exchanging and manipulating construals. In due course, such environments can be expected to provide a foundation for visionary frameworks for communication such as have been envisaged by Levy, in which forms of language and notions of creativity embracing both East and West traditions are represented.

The authors wish to thank Todd Lubart, Chen Jin, Steve Russ and Yun Pui Yung for ideas and background work that have contributed to this paper.

[1] Beynon, W.M., Empirical Modelling, Radical Empiricism and the nature of knowing, Pragmatics and Cognition, 13:3, 2005 (to appear)

[2] Beynon, W.M., Adzhiev, V.D., Cartwright, A.J., Yung, Y.P., A Computational Model for Multiagent Interaction in Concurrent Engineering, Proc. CEEDA'94, Bournemouth Univ., 227-232, 1994.

[3] Beynon, W.M., Adzhiev, V.D., Cartwright, A.J., Yung, Y.P., A New Computer-Based Tool for Conceptual Design, Proc. Workshop Computer Tools for Conceptual Design, Univ. of Lancaster, 171-188, 1994.

[4] Beynon, W.M., Adzhiev, V.D., Cartwright, A.J., Yung, Y.P., An Agent-oriented Framework for Concurrent Engineering, Proc. IEE Colloquium: Issues of Cooperative working in Concurrent Engineering, Digest 1994/177, 9/1-9/4, October 1994.

[5] Beynon, W.M., Rasmequan, S., Russ, S.B., A New Paradigm for Computer-Based Decision Support. Decision Support Systems, vol 33, 2002, 127-142.

[6] Beynon, W.M., Roe, C.P., Empirical Modelling principles to support learning in a cultural context, Proc. 1st International Conference on Educational Technology in Cultural Context, University of Joensuu, Finland, 2002, 151-172.

[7] Beynon, W.M., Roe, C.P., Computer support for constructionism in context. Proc. of ICALT'04, Joensuu, Finland, August 2004, 216-220.

[8] Cao, S., Klein, R., Zheng, G., Geng, W., Metis Global Virtual Network Toward Beijing Olympics 2008: Managing Uncertainty in Media Content Platforms. Total Innovation Management mediated by CSCW Design. Proc. 4th Intl. Symp. on Management of Technologies. Zhejiang UniversityPress, Hangzhou, 2004, 171-175.

[9] Cao S., Klein R., Liu J., Enhancing the usage pattern mining performance with temporal segmentation of QPop Increment in digital libraries. J Zhejiang Univ SCI 6A(11), 2005, 1290-1296

[10] Cartwright, R., Adzhiev, V., Pasko, A., Goto, Y., Kunii, T.L., Web-based shape modelling with HyperFun. IEEE Computer Graphics and Applications, 25(2), 2005, 60-9.

[12] Detienne, M., Vernant, J.P., Les ruses de l’Intelligence, la Metis des grecs. (Ed. Champ Flammarion), 1974.

[13] Fischer, Gerhard, Social Creativity: Making All Voices Heard, Proc. HCII 2005, Volume 10: Internationalization, Online Communities and Social Computing: Design and Evaluation, 2005

[14] Joas, H. , “La créativité de l’agir”, trad. De l’allemand par Pierre Rusch (éd. Cerf) Paris, 1999.

[15] Jullien F., “Procès ou Création, Une Introduction à la Pensée des lettrés chinois”. (Des travaux / Seuil), 1989.

[16] Klein R., ‘ La Metis-pour-créer ? ’Vers l’Analyse Médiologique d’une Métaphore: La Créativité Selon la lecture de l’ouvrage de François JULLIEN (1989) : “Procès ou Création” . Mémoire de DEA Sciences InfoCom. Université de Nice –Sophia-Antipolis, France . 2002, 223pp.

[17] Klein R., METIS: Chinese-Western Intercultural Approach to innovation and Global virtual teams CSCW for talent students and innovative individuals, International Forum on Education Model and Policy Issues for Cultivating Talent Student / P.1-4 , HangZhou PR China, 22- 23rd May, 2004

[18] Klein R., How to Manage Global Virtual R&D cross cultural Teams using a groupware in the film industry, Proceedings Beijing 2008 Olympic Games Symposium, HOSC, Renmin University of China. June 2004, at url: http://www.c2008.org/rendanews/english_te.asp?id=894 (accessed 10/11/2005)

[19] Klein R., Letaief R., Carter S., Chabert G., Lasonen J., Lubart T., CSCL for NGO's Cross Cultural Virtual Teams in Africa: An Ethiopian Children Advocacy Case Study against Exclusion and toward Facilitation of Expression, Innovation and Creativity . Proc. ETCC-ICALT'05 Kaohsiung, Taiwan, 2005, 1037-1041.

[20] Latour B., Petite réflexion sur le culte moderne des dieux faitiches, Paris, Ed. Les empêcheurs de penser en rond, 1996.

[21] Lévy, P., “Cyberculture: Rapport au Conseil de l’Europe.” (Ed. Odile Jacob). 1997

[22] Lévy-Leblond J-M., L’esprit de sel: Science, Culture, Politique. (Points/ Nouvelle édition/ sciences), 1984.

[23] Lubart, T.I., “Creativity across cultures” Handbook of creativity (R.J. Sternberg ed.) Cambridge Press, 1999.

[24] McCarty, W, Beynon, W.M., Russ, S.B., Human Computing: Modelling with Meaning, in ACH+ALLC 2005, 138-144

[25] Pang N., Cao S., Schauder D., Klein R., A Hybrid Approach in the Evaluation of Usability for Multimedia Objects: Case Study of the Media Assets Management Platform for an Advertainment Production Project toward Beijing Olympics 2008 . Proceedings ICITA 05 Sidney, 2005

[26] Tse E. Histon J., Scott S., Greenberg S., Avoiding Interference : How people use spatial separation and partitioning in SDG workspaces. CSCW 2004, Chicago Vol. 6 Issue 3, 2004, 252-261

[27] Xu L.,Cao S., Klein R., Hu B.,Chen J., Jin Y., METIS: Advertainment Movie for Beijing Olympics 2008: A Study of Cultural Economy and Cross-cultural Collaborative IT Innovation . Proceedings Beijing 2008 Olympic Games Symposium, HOSC, Renmin University of China. June, 2005, 54-68

[28] CIweb - Collective Intelligence and IEML - at url : http://137.122.100.152/