аЯрЁБс>ўџ CEўџџџBџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџьЅСk №П—AbjbjЎЎ %L}Ф}Ф—=џџџџџџlііііііі фффф № ФhCEEEEEE$, LЮiіiDіі~DDDііCDCD:D~ їііC `од’У кф.C”0Ф,DCD  іііійComputing for Construal: Making Sense and Making Science 1. Context Two kinds of model-building activity are involved in the comprehension of complex systems and phenomena. One of these takes place 'post-theory', when reliable knowledge and understanding has been attained. The other occurs 'pre-theory', before reliable observations and interpretations have been identified. An example of a pretheory model is the mechanism that Faraday (1831) built to reveal that the ‘rotating discs’ observed in organisms by Leeuwenhoek (1702) were an optical illusion. Post-theory models are typically mathematical, framed using abstract representations and logically consistent propositions, and current computer science is well-oriented towards their implementation. Pre-theory model-building is a much more untidy, personal matter, where knowledge is pre-articulate and boundaries are not clearly defined. Such models are hard to communicate; they often relate to a specific situation and to tacit knowledge that is only accessible through interaction; they may embrace ambiguity and conflict. They are nonetheless the essential personal starting points of scientific investigation. Theories are the points of arrival for our extensive sense-making journeys from such private construals to public expression. Such journeys are haphazard and unpredictable, and may also take place at many levels, as when one theory is adopted in the search for another. 2. Challenge and Approach The challenge that is addressed in this project is: exploiting the computer for pre-theory model-building, in making sense and making science. The target users for whom this challenge is significant include experimental scientists, participants in conceptual design and systems analysts; it has most relevance to human-centred activities, such as learning and decision-support, where craft takes precedence over theory. Computing has brought some benefits to these areas through technologies such as databases, spreadsheets, visualisation tools, active white boards etc, and through packaging knowledge and know-how in expert systems and problem-solving environments. Such resources comprise tools based on post-theory models, integrated through generic data representations. Where sense-making is concerned, human and computing activities are treated as separate concerns, and there are no coherent principles and tools for integrating model-building pre- and post-theory. The computer-based model-building approach that we exploit in this project (‘Empirical Modelling’) builds on existing computing concepts and tools, such as declarative programming and spreadsheets, but has a fundamentally different emphasis from the classical theory of computation. Rather than optimising and automating the behaviour of the computer for specific uses and goals, EM focuses on giving computer support to the human creation and maintenance of intelligible state. In EM, the commonsense notions of observation, dependency and agency that are prominent in experimental science take precedence over computational and logical abstractions. The special-purpose tools that we have developed for EM promote the construction of artefacts that embody patterns of observation, dependency and agency. Identifying such patterns, which are far less restrictive than the comprehensive behaviours to be preconceived in conventional programming, is the key activity for pre-theory model-building. Interaction with EM artefacts has qualities, characteristic of commonsense human agency, that are quite inadequately represented in traditional accounts of computer science. EM liberates the power of the computer for communication that is based - fundamentally rather than exclusively - on sharing artefacts rather than on language. This has immediate practical implications for integrating manual and automated activities that go far beyond what is conceived in the consideration of usability. The products of EM are as open to live interaction and interpretation as the artefacts we encounter moment-by-moment in our experience; at the discretion of the modeller, they can be configured for a specific behaviour and used like a traditional computer program, but in such a way that this behaviour can always be suspended and subverted through reinterpretation. In truly experimental interactions, the particularity of the context, the uncertainty of the outcome, and the potential conflicts in interpretation all play a key role. Since interaction with EM artefacts can accommodate the modeller's awareness of situation, ignorance and nonsense, it also offers the capacity for surprise, confusion and error that is of the essence in real experiment (‘the principle of SIN’). Particular models illustrate how EM can give computer support for construal in many different contexts. These include: an environment developed on-the-fly in a teaching context that seamlessly integrates pure relational algebra with several SQL-like translators and interpreters to expose the logical flaws in SQL; a laboratory for the construction of board games in which the impact of possible rules, geometries, and perceptual and cognitive skills can be exposed layer-by-layer; a distributed model of railway operation in the vicinity of the Clayton Tunnel in 1861, where the roles of signalmen and drivers can be enacted in such a way as to reconstruct the circumstances of a historic accident, and to adapt the environmental conditions and mechanical characteristics of trains, signals and telegraphs etc to assess causality and attribute responsibility. 3. Aims and Objectives The new challenge addressed by this project is the consolidation and extension of our previous sense-making model-building and its application to science. Our aims are: (a) to disseminate and strengthen the 'basic technology' of EM as it has been developed at the University of ЬЧаФTV over recent years (see http://www.dcs.warwick.ac.uk/modelling). Specific objectives are: establishing project work and teaching of EM and its applications outside ЬЧаФTV; publicising and distributing our tools primarily through workshops and appropriate web resources and stimulating their open source development; making closer links between EM and other disciplines (ranging from computing, engineering and physical sciences to philosophy, psychology and education). (b) to establish collaborations to exploit and further develop many aspects of computer support for construal for which proof-of-concept has been established in our previous research. Specific objectives: to pursue a collection of research projects designed to establish the use of our methods in science; to clarify links with existing approaches in areas such as end-user programming, scientific visualisation and problem-solving environments; to gain a fuller appreciation of the potential for EM in commercial application; to identify the methodological and technical issues to be addressed in its future development. (c) to consolidate our understanding of the 'basic science' of EM, and to defend and appraise critically its controversial stance with respect to the classical theory of computation and to received philosophies of science. Specific objectives are: developing publications and models, especially collaborative products; organising an international research workshop. 4. Significance and Impact Because of the pervasive role of computer-related technology, our new approach to model-building can have an enormous and wide-ranging impact. Potential applications include: enhanced support for experimental science, especially in areas such as medicine and biology, where mathematical models are limited; well-founded modelling for virtual reality environments such as are developed in engineering design and in the simulation of interactive processes; and aspects of computing where empirical techniques currently play a prominent role, such as performance analysis and tuning, requirements analysis, testing and informal knowledge representation. Wider applications of social and economic benefit include multi-variate decision support for business, educational and special needs technology, ubiquitous computing environments and advanced forms of non-linguistic computer-assisted communication with potential for conflict resolution. In the longer term, the project will contribute to a change in our current perception of the essential character of computer science with radical implications for software development and for challenging issues such as the scope for discovery in ‘virtual science’. 5. People, Methodology and Organisation At the core of the project will be the two directors of the EM project, Beynon and Russ, who will work in conjunction with Nudd as an advisor experienced in the management of large research projects, Allen as a scientific advisor, and Joy as a technical consultant. The project will require a Project Coordinator, a Technical Support expert and a Project Manager. (It may also be necessary to relieve key project members from teaching responsibilities in some phases of the project.) We anticipate being able to appoint a dedicated coordinator who has enough knowledge and experience of EM to supervise research. The technical support required will include development and maintenance of the EM tools and website, and responsibility for organising a technical consultancy helpline. In addition to the core members and the personnel directly involved in supervision and consultation, the project will have advisors from the ЬЧаФTV Centres for Scientific Computing and for New Technologies for Research in Education, as well as several interested international interdisciplinary collaborators (Pasko - Computer Science - Japan, Sonnenwald - Information Science - USA, Fischer - Engineering - Brazil, Suttinen - Education - Finland, Petsche - Philosophy - Germany). It will also have industrial advisors, to include Richard Cartwright at the BBC R&D labs, who has pioneered the application of EM principles to otherwise seemingly intractable problems in broadcasting. The EM group at ЬЧаФTV has trained many research students in a wide range of application areas. It is unusual in that it has benefited enormously from the enthusiastic contributions of undergraduate students, even in respect of research topics. A distinctive feature of the research output is the large body of diverse and richly interconnected models that has grown up over the last fifteen years; this collection of models is the primary resource for the entire enterprise of ‘Computing for Construal’. There are many significant models apart from the 70 or so that are currently in our public archive, and this number is growing ever more rapidly year-by-year as more students are being exposed to EM through final year projects and our 30 hour 4th year introductory module. The time required to archive a typical EM project so that it is accessible for general use and re-use can vary from hours to days. To meet the demands of maintaining the archive, and make the routine extensions by way of case-studies and new notations needed to support the core EM work at ЬЧаФTV, we intend to employ small teams of EM-literate students in summer vacations. We hope that dissemination will enable EM research groups with similar vitality to be established at other UK universities, and in due course lead to more interest in the collaborative development of our open source software tools. Our experience at ЬЧаФTV suggests that a modest introduction to the EM principles and tools is sufficient to initiate independent model-building work from able students, but that communication through personal demonstration and commentary has a vital role in their initial orientation. We have also observed that practical experience of EM is an essential prerequisite to engagement with the underlying principles, and it is on this basis that our dissemination strategy will focus initially on practical collaboration. There is evidence that reuse and incremental modification of EM models over the Web is a powerful novel paradigm for concurrent development, and this will itself contribute to the research profile of the project. It also gives scope for the involvement of industrial partners to exploit our principles and tools. The project will run over four years: this will comprise a period for initial training and start-up activities (1/049/04), a three year programme for core research (10/04-9/07) and a period for consolidation and evaluation (10/07-12/07). The core research activity will comprise nine collaborative projects aimed at the general theme of supplying computer support for construal in scientific applications. These will be organised into three strands: Cognitive aspects [Russ] Burke Operational Research Nottingham modelling dynamic situations, decision support Gooding Philosophy of Science Bath modelling experimental activity in science Kutar User Psychology Hertfordshire human factors, usability issues From experiment to theory [Project Coordinator - TBA] Bhalerao Image Processing ЬЧаФTV visualisation for medicine, experiential aspects Nehaniv Adaptive Systems Hertfordshire evolution and emergence in theory-building Sinclair Formal Methods ЬЧаФTV from pre-articulate to formal aspects Agency and explanation [Beynon] Burd Software Engineering Durham model comprehension and explanatory models McCarty Humanities Computing KCL sense and science by modelling and experiment Newell Assistive Technology Dundee personalised modelling, different perceptions A research assistant with training in EM will be assigned to each project. Each RA will work under the supervision of a core team member at ЬЧаФTV and will liaise with an academic acting in the role of a consultant from one of the collaborating research groups identified in the above table. An important subsidiary function for the RAs will be to promote interest in the practical adoption of EM (e.g. in project work, or small case studies within an existing module), and we shall stimulate this in the early stages of the project by holding short workshops on EM at several partner sites. There will be annual project management workshops for the entire project team, at which particular attention will be given to the alignment of the individual research projects within each strand, and in relation to the overall project theme. Regular seminars will be organised at ЬЧаФTV to address topical subthemes that may reflect the above strands or orthogonal technical issues (such as relate to software development). Potential industrial partners would also be invited. In the latter stages of the project, when the basic technology transfer has been effected, we shall organise two research workshops to disseminate our project findings and address their broader and potentially more controversial implications. The first will be a User Forum in the summer of 2005, where we will look to identify scientists who can use our tools directly, join in existing collaborations, or establish further independently funded projects. The second will be a final workshop in the summer of 2007 to which we intend to attract such interested parties as our worldwide EM collaborators and members of the Cognitive Technology Society. To this end, we propose to organise our workshop in conjunction with a sequel to the biennial international Cognitive Technology conference organised by Beynon, Russ and Nehaniv at ЬЧаФTV in 2001. At that point, we shall be in a position to assess how much progress we have made from Faraday's visual and mechanical model-building towards using computing technology to deliver the aspiration implicit in the title of CT2001: ‘Instruments of Mind’. 9DЅ П ѓ L ЪгепуыЬуŒ>Г Ю ‚%Њ%d+4—A§ёшёшмшмшмшмшёшЫшёшёшш B*CJOJQJ\^JaJph6B*CJ]aJphB*CJaJph5B*CJ\aJphCJ9DЅ П  roЬуŒзEГ Ю ‚%Њ%d+4е5ю5ѓссссссссесссссссЭсГ & F Ц8„а„8Єd1$7$8$H$]„а^„8„аЄd^„а „а„аЄd]„а^„а„а„аЄd1$7$8$H$]„а^„а „а„аЄd]„а^„а—Aўю5G6˜6п67j7О78'8v8Ш89K=—AщщщЯщщщЯщщщНН„а„аЄd1$7$8$H$]„а^„а & F Ц8„а„8Єd1$7$8$H$]„а^„8„а„а„аЄ(1$7$8$H$]„а^„а`„а #0PАа/ Ар=!Аh"Аh#а$5%А i8@ёџ8 NormalCJ_HaJmH sH tH \@\ Heading 1$$„„1$7$8$@&H$]„^„5B*\aJph<A@ђџЁ< Default Paragraph FontJY@ђJ Document Map-D MЦ џ€ OJQJ^JVT@V Block Text„„1$7$8$H$]„^„B*CJaJphLB@L Body Text„а1$7$8$H$]„аB*CJaJph—=Lџџџџ9DЅП r oЬуŒзEГЮ‚!Њ!d'0е1ю1G2˜2п23j3О34'4v4Ш45K9™=0€€˜0€˜0€€˜0€˜0€˜0€€˜0€€˜0€˜0€˜0€€˜0€˜0€€˜0€€˜0€˜0€€˜0€˜0€€˜0€€˜0€€˜ 0€˜0€˜0€˜0€˜ 0€€˜0€€˜0€€˜0€€˜ 0€€˜0€€˜0€€˜0€€˜0€€˜0€€—A"ю5—A#%—A$ єў™ ђ!ј!%")"Ў$Ж$я%є%&&Y&a&y&€&˜2233j3q34%4'4+4а4й4Й5П566 8Љ8o<u<€<‡<™=ГТю&ћ&™=333џџ Meurig Beynon*C:\My Documents\BasicTech\FINAL\FINAL3.doc Meurig BeynonJC:\WINDOWS\Application Data\Microsoft\Word\AutoRecovery save of FINAL3.asd Meurig BeynonJC:\WINDOWS\Application Data\Microsoft\Word\AutoRecovery save of FINAL3.asd Meurig BeynonJC:\WINDOWS\Application Data\Microsoft\Word\AutoRecovery save of FINAL3.asd Meurig BeynonJC:\WINDOWS\Application Data\Microsoft\Word\AutoRecovery save of FINAL3.asd Meurig Beynon*C:\My Documents\BasicTech\FINAL\FINAL3.doc Meurig BeynonJC:\WINDOWS\Application Data\Microsoft\Word\AutoRecovery save of FINAL3.asd Meurig BeynonJC:\WINDOWS\Application Data\Microsoft\Word\AutoRecovery save of FINAL3.asd Meurig BeynonJC:\WINDOWS\Application Data\Microsoft\Word\AutoRecovery save of FINAL3.asd Meurig Beynon*C:\My Documents\BasicTech\FINAL\FINAL3.doc›.ЃFя<>џџџџџџџџџh „„˜ўЦ^„`„˜ўOJQJo(З№h „и „˜ўЦи ^„и `„˜ўOJQJo(oh „Ј „˜ўЦЈ ^„Ј `„˜ўOJQJo(Ї№h „x„˜ўЦx^„x`„˜ўOJQJo(З№h „H„˜ўЦH^„H`„˜ўOJQJo(oh „„˜ўЦ^„`„˜ўOJQJo(Ї№h „ш„˜ўЦш^„ш`„˜ўOJQJo(З№h „И„˜ўЦИ^„И`„˜ўOJQJo(oh „ˆ„˜ўЦˆ^„ˆ`„˜ўOJQJo(Ї№›.ЃFџџџџџџџџ         џ@€66œd––66@„Р{—=А@џџUnknownџџџџџџџџџџџџG‡:џTimes New Roman5€Symbol3& ‡:џArial?5 ‡:џCourier New5& ‡:џTahoma;€Wingdings"Aˆ№аh u†ШuІuІСшЩ2l$№ЅРxxƒ2d^>2ƒQ№пџџ8Computing for construal: making sense and making science Meurig Beynon Meurig Beynonўџ р…ŸђљOhЋ‘+'Гй0Ф˜ ф№ 4 LX t € Œ ˜ЄЌДМф9Computing for construal: making sense and making scienceeompMeurig Beynoncoeureur Normal.dotoMeurig Beynonco28rMicrosoft Word 9.0r@Ц7і@6љ‹У@Юb[У@@эЕ’УшЩ2ўџ еЭеœ.“—+,љЎ00 hpŒ”œЄ ЌДМФ Ь фЬЧаФTV Universityl^>2 9Computing for construal: making sense and making science Title  !"#$%&ўџџџ()*+,-./01ўџџџ3456789ўџџџ;<=>?@Aўџџџ§џџџDўџџџўџџџўџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџRoot Entryџџџџџџџџ РFјд’УF€1Tableџџџџџџџџџџџџ'WordDocumentџџџџџџџџ%LSummaryInformation(џџџџ2DocumentSummaryInformation8џџџџџџџџџџџџ:CompObjџџџџjObjectPoolџџџџџџџџџџџџјд’Ујд’Уџџџџџџџџџџџџўџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџџўџ џџџџ РFMicrosoft Word Document MSWordDocWord.Document.8є9Вq