Paul, 'COMPUTER-BASED SIMULATION MODELS FOR PROBLEM-SOLVING: COMMUNICATING PROBLEM UNDERSTANDINGS', Arachnet Electronic Journal on Virtual Culture v2n02 (May 16, 1994) URL = http://hegel.lib.ncsu.edu/stacks/serials/aejvc/aejvc-v2n02-paul-computerbased The Arachnet Electronic Journal on Virtual Culture __________________________________________________________________ ISSN 1068-5723 May 16, 1994 Volume 2 Issue 2 PAUL V2N2 ======================================================================== COMPUTER-BASED SIMULATION MODELS FOR PROBLEM-SOLVING: COMMUNICATING PROBLEM UNDERSTANDINGS Ray J. Paul Brunel University Ray.Paul@brunel.ac.uk Peter J. Thomas University of the West of England Peter.Thomas@pat.uwe.ac.uk Abstract A recognition that computer-based tools communicate mental models among developers, users and other stakeholders, thereby placing constraints on the use of those tools, is important in understanding the use of technology. In our work in computer- based simulation modelling, realization that technology can both impose constraints and provide possibilities has been a constant concern. Computer-based simulation models aid the process of formulating a problem. Such models can act as "dynamic intermediaries" which facilitate the ebb and flow of understanding among stakeholders. This facilitation effect leads to a particular view of the role of technology in decision support. It has enabled us to develop software which improves the process of communicating problem understandings. This paper describes the use and theoretical foundations of computer-based tools we have developed to support simulation modelling. 1.0 Introduction The concept of "mental model" has acquired currency in research literature on the design of computer systems. The notion of a mental model as a "psychological working model of a system" (Johnson-Laird, 1983) is one of the cornerstones of human- computer interaction (HCI) research and of user interface design [1]. Studies have described possible types of mental model (Du Boulay, 1981), strategies for eliciting models from users (Halasz and Moran, 1983; Young, 1981), how such models can be tailored and adapted (Carroll and Carrithers, 1984), and the reasoning strategies on which models are based (Riley, 1986). In user interface design, the notion of mental models has been of practical use. Norman's (1983) distinctions among "design model", "user model", and "system image" have enabled interface designers to draw systematically on users' mental models (Bewlay et al, 1983). By paying careful attention to the conceptual model presented to the user (including manuals, training, and documentation as part of the system image), a user interface can selectively encourage the development of more accurate and more complete user models of the system. 1.1 Problems With "Mental Models" Several points about this formulation of "mental models" are worth noting. First, even though the power of constraints imposed by the design model has been recognised, there has been little research into the way in which the HCI design process can be manipulated systematically to good effect -- despite voluminous research [2]. For example, recent approaches such as Design Rationale (MacLean et al, 1990) tend to focus on structuring the process of design. Studies which aim to illuminate the designer- user relationship only address the ways in which users' "situated understandings" are structured (Suchman 1987). Second, the notion of "mental model" is defined typically in individualistic terms. In this view, mental models are employed by individuals (designers or users) as a way of understanding actions, events and objects. This contradicts a growing understanding that users' understandings of systems and their use is "socially constructed" [3]. For example, accounts of the implementation failures of large-scale systems (Paul 1993) have demonstrated that the influence of users, designers and stakeholders -- and the constraints under which systems are developed -- are far more influential than the individual user's understanding of a system. Third, the usual definition of mental model is predominantly mechanistic. This is to be expected, due to HCI's foundation in information processing models of cognition (Card et al, 1983) -- despite arguments which suggest that creating sufficiently powerful, generalisable and useful theories on this basis is problematic (Landauer, 1983). Fourth, one of the defining features of much work in HCI --that human-computer interaction is a matter of "interacting" with computers (Nickerson, 1976) -- has suggested that a user's mental model is built in response to the actions of an interactive partner (Luff et al, 1991; Norman and Thomas, 1991). This approach runs counter to suggestions that human-computer interaction is fundamentally a "media process" (Bodker 1991). And, finally, despite the growing body of work in disciplines such as Computer-Supported Cooperative Work (CSCW), the focus in system and application design still is predominantly on the individual user of a single system. The broader contexts in which users work is largely regarded as of secondary importance. However, the fact that there are many stakeholders clearly is important in many technologies. In the case of computer-based simulation modeling, many applications are large-scale commercial ventures in which the various agenda of stakeholders embody competing views on how to employ expensive resources. Tension exists between those competing agenda and the need to arrive at a solution which will satisfy all stakeholders' interests to an acceptable extent. For example, El Sheikh et al (1987) report on the use of a simulation model in planning future berth requirements for a Third World shipping port. Stakeholders consisted of the port authority, the government, an international funding agency, a company of consulting engineers, and an operations research specialist. The port authority required that the new port be adequate to handle their estimate of berth requirements; the government -- anxious to secure funding -- required that the development be as large as possible; the World Bank, who were to fund the development, were concerned that expenditure on construction work be minimised; the consulting engineers were anxious to secure a contract for the development of the port. This example demonstrates the power of computer-based simulation for handling a potentially difficult situation. Although the results of the simulation were reasonably well known, the model's benefits derived from the discussion that took place around the results: Participants suggested that their viewpoints might be different if parameters changes were made. The simulation model made rapid testing of such parameter changes possible, with the result that some if not all of the suggestions made were in fact infeasible. 1.2 Enriching the Concept of "Mental Model" As a consequence of these observations, it seems to us that a richer formulation of "mental model" encompasses the way in which computer technology is embedded in complex contexts, stakeholder (designer and user) activities, and tasks. A corollary to this need for reformulation is that technology is partly a "communication medium for understandings". For example, the development of applications such as CASE tools involves a mapping of tool designers' mental models of design processes onto the behaviour of tool users. Similarly, the development of "productivity applications" such as word-processors, diarisers and personal information managers (Thomas, forthcoming) involves the mapping of designers' understandings of those tasks onto the ways in which such tasks are to be performed by users. Of course, from one perspective the issue is trivial, in that all design -- from art, through architecture, to the design of everyday artifacts -- involves some sort of mapping between design intent and eventual use. What is more interesting is how to direct and exploit the relationships among designers, artifacts and users so that systems are made more usable. In the case of our work in computer-based simulation modelling, a recognition of the way in which technology can both impose constraints and provide possibilities has been a constant concern. The work that computer-based simulation models can do (aiding the process of formulating complex problems) rests on the assumption that computer-based simulation models act as "dynamic intermediaries" which facilitate the ebb and flow of various kinds of understandings between stakeholders. A computer-based simulation model then is explicitly seen as a tool to allow understandings of a problem situation to develop, to be discussed, and to be refined toward a solution. 1.3 Applications in Ongoing Simulation Modelling Work These examples show how computer technology (in the form of simulation models) has led us to a particular view of the role of technology in decision support. It also has enabled us to develop software which we suggest improves the process of communicating problem understandings. In the following sections, we look at some of the aims of computer-based simulation modelling as developed by the Centre for Applied Simulation Modelling (CASM) at the London School of Economics, and at Brunel University. At the outset, it should be recognised that the process of problem communication itself and the nature of mental models are not areas in which we have had particular empirical research interests. Rather, we are interested as researchers in application techniques in computer-based simulation modelling -- in ways that the process of simulation modelling can be facilitated through the development of more sophisticated technology to communicate problem understandings. 2.0 Computer-based Simulation Modelling 2.1 The Traditional View of the Modelling Process Simulation modelling is particularly popular amongst the operational research and information systems fraternities. Practitioners would offer the following description of the process of simulation modelling. o There is a real world problem. o This problem is formulated as a logical model in the form of activity cycle diagrams, flow charts, or block diagrams (amongst others). There are a variety of ways to represent the logic of a formulated problem. o The logical model is converted into a computer model (sometimes a computer program, other times a data driven generic simulation system). o This computer model is verified (tested to see if it is doing what the analyst wants it to do). o The model then is used as an operational model to produce some results or some conclusions, or for implementation (after the operational model has been validated against the real world). An implicit assumption in this process is that the product of modelling is a set of results (usually numerical) which lead decision makers and/or analysts to some conclusions, from which some decisions are implemented. 2.2 Our View of the Modelling Process However, our view of the process is somewhat different. First, in many real world situations problems are owned by interest groups (some of whom are in conflict). The definition of the problem is influenced by the owners of the problem -- especially in complex strategic decision making. Second, because the problem is complex, formulation of it is a difficult task. The construction of a logical model representing the formulation of the problem is, in many instances, the most difficult aspect of the problem treatment. Third, understanding what the problem is may be the object of the whole exercise. This means that an analyst should be prepared to undertake problem reformulation constantly, to obtain a common understanding of the problem as part of the modelling process. And, finally, a dynamic logical model needs to be turned into a computer model with relative ease. If this part of the process takes a long time, contact with the real world problem starts to diminish. For example, if the analyst discusses the computer model with the decision-makers infrequently, then the chance that the computer model will represent the real world problem is small. In many instances then, the computer model serves as a medium of communication of problem understandings, for all participants in the decision making process. It obviously is necessary to verify that the computer model does what one thinks it should. But it is questionable how much emphasis should be placed on generating an operational model to be used for experimentation purposes. In many cases, the production of a computer model which secures problem definition agreement among the decision-makers may be sufficient to satisfy all participants. 2.3 Major Issues and Problems There are a number of issues and problems associated with using simulation modelling as a decision aiding technique which we have attempted to address, partly based on our understanding of simulation models as a medium of communication. First, most problems to which simulation applies are poorly defined. In fact, if the problem is not poorly defined there probably are better and more reliable methods of solving it than the rather crude technique of simulation modelling. Second, any complex, important problem probably will involve conflicting interests and understandings. If the modelling process is going to lead to change, it is unlikely that all decision makers will see these changes as favourable to them. As much as possible, the modelling process should be used in a neutral way to help the participants in decision-making understand their problem. Third, the specification of "the problem" is never static. Even if one succeeds in satisfying the conflicting views of the decision makers, for complex problems it is probable that the specification will still undergo change. The real world is dynamic. Therefore, the perceived problem will be dynamic, as well. The fourth problem is the question of "model confidence". No computer simulation of any size can possibly be verified, and no model of any size can possibly be validated against the real world, given that the real world is not static. A final and crucial feature of simulation modelling is that it involves "decision aiding". Discrete event simulation modelling is a quantitative technique. The outputs are numerical, and numerical values tend to indicate that one course of action might be better than another. However, such numerical techniques cannot represent all possible factors in the problem scenario. They can represent crudely most or some factors in a quantitative way, but they cannot represent subjective factors. In our view then, the simulation modelling process is not designed to find "the answer". Rather, it aims to help decision- makers make decisions or to gain an understanding of a problem. It follows, therefore, that the development of computer systems for simulation modelling is driven by the need to make simulation efficient as a modelling tool, for helping decision makers understand their problems. Therefore, some prime concerns are o to recognise the role of such tools in communicating understandings, o to provide methods by which computer-based tools can communicate those understandings effectively, and o to design tools so that analysts can develop simulations which embody problem understandings. The following section discusses in greater detail the kinds of tools which we have developed to fulfil these requirements. 3.0 Computer-based Environments for Simulation Modelling The essential feature of simulation modelling -- problem formulation and understanding -- has led us to develop computer- based tools which facilitate their role as dynamic intermediaries. 3.1 Areas of Development We have undertaken development in several areas: specification methods, problem formulators, and interactive visual simulations. It is this last area which we suggest is most effective in communicating problem understandings through technology. We briefly review all three to highlight the differences among them. 3.1.1 Specification Methods An essential part of simulation modelling is specifying the problem. If a specification is going to be used as a vehicle for communication, it must have a simple structure. However, many simulation models inherently model complex situations: the combination of objects or entities in an activity requires some complex conditions to be stated. If these conditions are described explicitly in the specification method, then the specification becomes difficult to follow. There are several ways the problem can be specified, such as diagrammatic techniques (Ceric and Paul, 1989) -- including activity cycle diagrams and Petri nets, and semi-formal or mathematical methods (Zeigler, 1984). Paul and Ceric (1992) propose the principle of "Comprehensive Harmony" as a requirement for the specification method: the method must be reasonably comprehensive. However, comprehensiveness must be balanced by a harmony in the method that makes it intelligible to the active participants in the simulation modelling process. 3.1.2 Problem Formulators A problem formulator (Balmer and Paul, 1986) assists the analyst and decision-maker in formulating the nature of a problem. The system captures the model logic of the problem, then applies an interactive simulation program generator (ISPG). The ISPG produces a simulation model which calls a library of software subsystems to run the simulation itself. Simulation model output is analysed by an output analyser, which would help determine experimental designs for running and controlling the simulation model. The problem formulator and output analyser close the loop, so that the analyst and decision-makers collectively use the complete system. Attempts have been made by CASM to develop a problem formulator (Doukidis, 1985 and1987; Doukidis and Paul (1987); Paul and Doukidis, 1986; Paul, 1987). 3.1.3 Interactive Visual Simulation In an attempt to overcome these disadvantages, CASM has developed tools for interactive visual simulation. These provide a dynamic and highly-usable prototyping environment in which a problem owner and analyst can build a simulation model in a collaborative manner. Visual modelling is a powerful component of an analyst's problem- solving capabilities. Existing specification methods normally require a translation process between abstract psychological representation (the analyst's problem understanding) and the logical representation of the model (in terms of diagramming techniques or macro languages, for example). Interactive visual simulation tools, on the other hand, provide a simulation environment which enables constant reconstruction of the model -- by allowing the user to draw a visual representation of the real world system. 3.2 Macintosh Graphical Simulation Environment (MacGraSE) A research system (MacGraSE) [5] based on these observations and experiences was developed for the Apple Macintosh. The development aims to investigate the use of an environment which allows users to create a visual version of their understanding of the problem directly (during construction of the model), without being required first to translate the modeller's understandings into a logical structured form. We briefly shall describe some features of the system. It is clear that providing context for the decision-maker and the analyst is an important part of formulating problems and communicating understanding using such a system. One way in which this can be accomplished is through the use of simple visual elements, such as "fixed backgrounds" against which entities modelled in the system can be represented. A fixed background might take the form of a picture of the real-world problem being modelled (the layout of a port, of a production line, or of an out-patient clinic, for example). On the surface of the fixed background, graphical objects are used to represent entities in the simulation. In order to make the pictorial descriptions represent the real image of the system, each entity type has its own iconic representation in the model. The model logic then can be defined by using the icons on the screen. Icons indicate Entities (any component of the model which can be imagined to retain its identity through time), and Activities (active states in which Entities of different types engage in cooperative tasks of some duration). The display of Queues (passive states of an Entity type while it waits for an Activity to commence) and Attributes (values or characteristics of each Entity) are optional. The way that Entities move about in the system forms the basic model logic. The main advantage of using an application such as MacGraSE is its visual interface, which allows the user to have complete control over the application during construction of the simulation model. Moreover, the formulation mechanism allows the user to reconstruct the model continuously, throughout the model- building process. The tight indexing of Entity icons is so apparent that the user can modify an existing path or create an additional path easily anywhere within their life cycle. For simulation models that are heavily attribute-based, the user can model movements of attribute evaluation easily. MacGraSE was prototyped with a facility to run the model interpretively, so that the analyst and the customer could see not only the visual image of the problem being modeled, but also the outcomes of current model construction. This facility is essential if the objective is to move as soon as possible towards a tight specification of the problem being modeled. This iterative derivation of a model is, in its turn, highly dependent on the speed with which the model can be adapted to meet new understandings. The prototype demonstrates that the visual approach is capable of the required speed. However, it is yet to be determined whether or not the level of functionality required of a production version would be greater. Increased functionality might slow the tools' adaptation speed. The appropriate level of functionality is a balance between the need to make the system comprehensive enough for a wide range of problems, and the need to keep the system manageable. The existence of this tradeoff is one reason why we elected to incorporate a program generator into the prototype. The interpreted version of the model can handle the basic model structure, and allow for quick and dirty understanding. This first-cut model then can be expanded for the code-generated version to include the extra detail considered necessary to advance use of the model. It is a positive side benefit that this style of model construction encourages the analyst to "grow the model whilst retaining control of the understanding that is derived from the model. 3. CONCLUSION Although we have provided only a brief overview of processes involved in computer-based simulation modelling, it is possible to see that we conceive of the use of computer-based simulation models to be essentially one of applying technology to communicate problem understandings. Our experience is that using tools such as MacGraSE (and other visual simulators) enables us to focus explicitly on the communication of problem understandings, in such a way that problems become more open for discussion -- allowing easier resolution. Given that we have not been interested in this process primarily as a empirical phenomenon, we have little hard data on the detailed effectiveness of the computer tools we have developed. However, we suggest that studies of the use of systems such as MacGraSE (and of the use of computer-based simulation models in general) will support the assertion that such tools can support the process of decision making effectively, by acting as dynamic intermediaries. In terms of research into mental models and their role in computer technology, we further suggest that computer-based simulation modelling represents a useful "laboratory" -- a laboratory in which we can seek to develop more comprehensive understandings of the ways in which mental models are used in real-world settings. Footnotes 1. Carroll and Olsen, 1988; Briggs, 1988; Carroll and Rosson, 1987; Fischer, 1991; Lewis, 1986; Norman, 1987 and 1983; Gentner and Stevens, 1983 2. Bellotti, 1988 and 1990; Curtis et al, 1988; Dagwell and Weber, 1983; Gould and Lewis, 1985; Gould et al, 1987; Hammond et al, 1983; Farooq and Dominick, 1988; Johnson and Nicolosi, 1990 3. Heath and Luff 1991; Winograd 1987; Lawrence et al, forthcoming; Randall and Hughes, forthcoming 4. Other applications of computer-based simulations show the same characteristics, in domains such as reorganising the design of out-patient clinics in the health service (Kuljis et al, 1990), in stock control (Paul, 1989), in Flexible Manufacturing Systems developments (Hlupic and Paul, 1991, 1992a, 1992b, and 1994), and in military applications (Crookes et al, 1986; Holder and Gittins, 1989; Williams et al, 1989). 5. MacGraSe is primarily a demonstrator for the approach to visual interactive simulation and is available free of charge from CASM. Please contact Ray Paul (Ray.Paul@brunel.ac.uk) References Balmer, D. W. & Paul, R. J. (1986). CASM -- The right environment for simulation. Journal of the Operational Research Society 37(5), 443-452. Bellotti, V. (1990). A framework for assessing applicability of HCI techniques. In D. Diaper et al (Eds.), Proceedings of the Third IFIP TC13 Conference on Human-Computer Interaction Interact'90, 213-218. Amsterdam: North-Holland. Bellotti, V. (1988). Implications of current design practice for the use of HCI techniques. In D.M. Jones and R. Winder (Eds.), People and Computers IV, 13-34. Cambridge, UK: Cambridge University Press. Bewlay, W. L., Roberts, T. L., Schroit, D. & Verplank, W. (1983). Human factors in the design of Xerox 8010 Star office workstation. Human Factors in Computing Systems, Proceedings of the CHI'83 Conference (Boston), 72-77. New York: Association for Computing Machinery. Bodker, S. (1991). Through the Interface. (Hillsdale, N.J.: Lawrence Erlbaum Associates. Briggs, P. (1988). What we know and what we need to know: the user model versus the user's model in human-computer interaction. Behaviour and Information Technology 7(4), 431-442. Card, S. K., Moran, T. P. & Newell, A. (1983). The Psychology of Human-Computer Interaction. Hillsdale NJ: Lawrence Erlbaum Associates. Carroll, J. M. & Carrithers C. (1984). Training wheels in a user interface. Communications of the ACM 27(8), 800-806. Carroll, J. M. & Olsen, J. R. (1988). Mental models in human- computer interaction. In. M. Helander (Ed.), Handbook of Human-Computer Interaction, 46-65. Amsterdam: Elsevier. Carroll, J. M. & Rosson, M. B. (1987). The Paradox of the active user. In J.M. Carroll (Ed.), Interfacing Thought, 80-111. Cambridge, Mass.: MIT Press. Ceric, V. & Paul, R. J. (1989). Preliminary investigations into simulation model representation. In Proceedings, 11th International Symposium on "Computer at the University" (Cavtat, Yugoslavia, June). Crookes, J. G., Balmer, D. W., Chew, S. T. & Paul, R. J. (1986). A three phase simulation system written in Pascal. Journal of the Operational Research Society 37(6), 603-618. Curtis, B., Krasner, H. & Iscoe, N. (1988). A field study of the software design process for large systems. Communications of the ACM 31(11), 1268-1287. Dagwell, R. & Weber, W. (1983). System designers' user models: A comparative study and methodological critique. Communications of the ACM 26, 987-997. Doukidis, G. I. (1987). An anthology on the homology of simulation with artificial intelligence. Journal of the Operational Research Society 38(8)(August), 701-712. Doukidis, G. I. (1985). Discrete Event Simulation Model Formulation Using Natural Language Understanding Systems. Unpublished Ph.D thesis. (University of London, England). Doukidis, G. I. and Paul, R. J. (1987). Artificial intelligence aids in discrete event digital simulation modelling. In IEE Proceedings 134, Pt.D(4)(July), 278-286. DuBoulay, B., O'Shea, T. & Monk, J. (1981). The black box inside the glass box: Presenting computing concepts to novices. International Journal of Man-Machine Studies 14, 237-249. El-Sheikh, A., Paul, R. J., Harding, A. S. & Balmer, D. W. (1987). A microcomputer-based simulation study of a port. Journal of the Operational Research Society 38(8), 673-681. Farooq, M. U. & Dominick, W. D. (1988). A survey of formal tools and models for developing user interfaces. International Journal of Man-Machine Studies 29, 479-496. Fischer, G. (1991) The importance of models in making complex systems comprehensible. In M. J. Tauber & D. Ackerman (Eds), Mental Models and Human-Computer Interaction 2. Amsterdam: North-Holland. Gentner, D. & Stevens, A. L. (Eds.)(1983). Mental Models. Hillsdale, N.J.: Lawrence Erlbaum Associates. Gould, J. D. & Lewis, C. (1985). Designing for usability -- key principles and what designers think. Communications of the ACM 28, 300-311. Gould, J. D., Boies, S. J., Levy, S., Richards, J. T. & Schoonard, J. (1987). The 1984 Olympic Games messaging system: A test of behavioural principles of system design. Communications of the ACM 30, 785-796. Halasz, F. G. & Moran, T. P. (1983). Mental models and problem- solving in using a calculator. Human Factors in Computing Systems, Proceedings of the CHI'83 Conference, Boston, 212- 216. New York: Association for Computing Machinery. Hammond, N., Jorgensen, A., Maclean, A., Barnard, P. & Long, J. (1983). Design practice and interface usability: Evidence from interviews with designers. Human Factors in Computing Systems, Proceedings of the CHI'83 Conference (Boston, December), 40- 44. New York: Association for Computing Machinery. Heath, C. & Luff, P. (1991). Disembodied conduct: communication through video in a multi-media office environment. Human Factors in Computing Systems, Proceedings of the CHI'91 Conference, 99-103. New York: Association for Computing Machinery. Hlupic, V. & Paul, R. J. (1991). A review of simulation modelling in FMS. Proceedings, XIII International Conference on Information Technology Interfaces (Cavtat, Yugoslavia, June). Hlupic, V. & Paul, R. J. (1994). Development of FMS simulation models using activity cycle diagrams. Accepted by the Journal of the Operational Research Society. Hlupic, V. & Paul, R. J. (1992a). Software packages for manufacturing simulation : A comparative study. Proceedings, XIV International Conference on Information Technology Interfaces (Pula, Croatia, September). Hlupic, V. & Paul, R. J. (1992b). FMS scheduling strategies using the simulation package SIMFACTORY II.5. Proceedings, 8th International Conference on CAD/CAM, Robotics, and Factories of the Future. (Place Edouard Branly, Metz, France, 17-19 August). Holder, R. D. & Gittins, R. P (1989). The effects of warship and replenishment ship attrition on war arsenal requirements. Journal of the Operational Research Society 40, 167-175. Johnson, P. & Nicolosi, E. (1990). Task-based user interface development tools. In D. Diaper et al (Eds.), Proceedings of the Third IFIP TC13 Conference on Human-Computer Interaction Interact'90, 383-387. Amsterdam: North-Holland. Johnson-Laird, P. N. (1983). Mental Models. Cambridge, UK: Cambridge University Press. Kuljis, J., Paul, R., Malin, J. H. & Thakar, S. (1990). Designing an out-patient clinic modelling package. Proceedings, 12th International Symposium on "Computer at the University" (Cavtat, Yugoslavia, June). Landauer, T. K. (1987). Relations between cognitive psychology and computer system design. In J. M. Carroll (Ed.), Interfacing Thought, 1-25. Cambridge, Mass.: MIT Press. Lang, K. N., Auld, R., & Lang, T. (1982). The goals and methods of computer-users. International Journal of Man-Machine Studies 17, 375-399. Lawrence, D., Atwood, M., Dews, S. & Turner, T. (forthcoming). Social interaction in the design and use of a workstation: Two contexts of interaction. In P. Thomas (Ed.), The Social and Interactional Dimensions of Human-Computer Interfaces. Cambridge, UK: Cambridge University Press. Lewis, C. (1986). Understanding what's happening in system interactions. In D. A. Norman & S. W. Draper (Eds.), User Centred System Design, 169-185. Hillsdale, N.J.: Lawrence Erlbaum Associates. Luff, P., Frohlich, D. & Gilbert, G. (Eds.)(1990). Computers and Conversation. London: Academic Press. McLean, A., Bellotti, V. & Young, R. (1990). What rationale is there in design. In D. Diaper et al (Eds.), Proceedings, Third IFIP TC13 Conference on Human-Computer Interaction Interact'90, 207-212. Amsterdam: North-Holland. Nickerson, R. S. (1976). On conversational interaction with computers. In R. M. Baecker & W. A. S. Buxton (Eds.)(1987), Readings in Human Computer Interaction, 681-693. Los Altos, Calif.: Morgan Kaufmann. Norman, D. A. (1988). The Psychology of Everyday Things. New York: Basic Books. Norman, D. A. (1987). Cognitive artifacts. In J. M. Carroll (Ed.), Interfacing Thought, 17-38. Cambridge, Mass.: MIT Press. Norman, D. A. (1983). Some observations on mental models. In D. Gentner & A. Stevens (Eds.), Mental Models. Hillsdale, N.J.: Lawrence Erlbaum Associates. Norman, M. A. & Thomas, P. J. (1990). The very idea: Informing HCI design from conversation analysis. In P. Luff et al (Eds.), Computers and Conversation, 51-65. London: Academic Press. Paul, R. J. (1993). Why users cannot get what they want. In Proceedings, the Do Users Get What They Want Conference (Centre for Research into Innovation, Culture and Technology, Brunel University). Paul, R. J. (1991). Recent developments in simulation modelling. Journal of the Operational Research Society 42(3), 217-226. Paul, R.J. (1989) Use of simulation to investigate stock control policies. In Proceedings, Technical Overview Symposium on Management Control Systems and Information Technology Implementation Issues of Inventory and Stock Control Systems. (London, June). Paul, R. J. (1987). A.I. and stochastic process simulation. In B. Phelps (Ed.), Interactions in Artificial Intelligence and Statistical Methods, 85-98. London: Gower Technical Press. Paul, R. J. & Ceric, V. (1992). Methods of Model Representation in Discrete Event Simulation: An Overview. Submitted to Transactions of The Society for Modelling and Computer Simulation. Paul, R. J. & Doukidis, G. I. (1986). Further developments in the use of artificial intelligence techniques which formulate simulation problems. Journal of the Operational Research Society 37(8)(August), 78?-810. Randall, D. & Hughes, J. (forthcoming). Working with customers: CSCW and office work. In P. Thomas (Ed.), The Social and Interactional Dimensions of Human-Computer Interfaces. Cambridge, UK: Cambridge University Press. Riley, M. (1986). User understanding. In D. A. Norman & S. W. Draper (Eds.), User Centred System Design, 157-169. Hillsdale, N.J.: Lawrence Earlbaum. Suchman, L. A. (1987). Plans and Situated Actions: The problem of Human-Computer Communication. Cambridge, UK: Cambridge University Press. Thomas, P. (Ed.)(forthcoming). Personal Information Management. London: Unicom Publications Ltd. Williams, T. M., Gittins, R. P. & Burke, D. M. (1989). Replenishment at sea. Journal of the Operational Research Society 40, 881-887. Winograd, T. (1987). A language/action perspective on the design of cooperative work. Human Computer Interaction 3, 3-30. Zeigler, B. P. (1984). Multifacetted Modelling and Discrete Event Modelling. London: Academic Press. _____ Articles and Sections of this issue of the _Electronic Journal on Virtual Culture_ may be retrieved via anonymous ftp to byrd.mu.wvnet.edu or via e-mail message addressed to LISTSERV@KENTVM or LISTSERV@KENTVM.KENT.EDU (instructions below) or GOPHER gopher.cic.net Papers may be submitted at anytime by email or send/file to: Ermel Stepp - Editor-in-Chief, _Electronic Journal on Virtual Culture_ M034050@MARSHALL.WVNET.EDU _________________________________ *Copyright Declaration* Copyright of articles published by Electronic Journal on Virtual Culture is held by the author of a given article. If an article is re-published elsewhere it must include a statement that it was originally published by Electronic Journal on Virtual Culture. The EJVC Editors reserve the right to maintain permanent archival copies of all submissions and to provide print copies to appropriate indexing services for for indexing and microforming. _________________________________ _________________________________ _THE ELECTRONIC JOURNAL ON VIRTUAL CULTURE_ ISSN 1068-5327 Ermel Stepp, Marshall University, Editor-in-Chief M034050@Marshall.wvnet.edu Diane (Di) Kovacs, Kent State University, Co-Editor DKOVACS@Kentvm.Kent.edu ____________________________ GOPHER Instructions ____________________________ GOPHER to gopher.cic.net 70 ____________________________ Anonymous FTP Instructions ____________________________ ftp byrd.mu.wvnet.edu login anonymous password: users' electronic address cd /pub/ejvc type EJVC.INDEX.FTP get filename (where filename = exact name of file in INDEX) quit LISTSERV Retrieval Instructions _______________________________ Send e-mail addressed to LISTSERV@KENTVM (Bitnet) or LISTSERV@KENTVM.KENT.EDU Leave the subject line empty. The message must read: GET EJVCV2N2 CONTENTS Use this file to identify particular articles or sections then send e-mail to LISTSERV@KENTVM or LISTSERV@KENTVM.KENT.EDU with the command: GET where is the name of the article or section (e.g., author name) and is the V#N# of that issue of EJVC