Affiliations: Professor of Architectural and Urban Computing, The Bartlett School of Graduate Studies, University College London

Abstract: One of the fundamental aims of system design is to make the user unaware of a system's presence. You only become aware of something when it has gone wrong. And yet there is a paradox here. In order to design a system that is transparent to the user we seem to need to be completely aware of it and of its deep structure. This paradox is associated with what is known as the 'problem of mapping'. This is the problem of how to move from abstract representations of tasks, knowledge and data structures to a concrete representation of the structure of dialogues and interactions required for a specific implementation. In this paper I review some of the history of a different field of design - architecture - in its search for design methodologies. Using that history I suggest that underlying the paradox lies a paradigm - a set of ideas that we think with rather than of - in which some of our most deeply held assumptions may be at fault. In particular I suggest that the established task modelling approach assumes as basic the distinction between user and system, while for any 'well working' system the user must actually become literally embedded. That is, the distinction between user and system should, from the user's point of view, effectively disappear. I suggest that this sort of disappearance can result from consideration of design in terms of designing and managing ecologies. I illustrate this with studies of health buildings. This suggestion seems to be in line with recent thinking in both philosophy and cognitive science, however it does raise serious questions for methodology in engineering design - by which I mean in the broadest sense, systems which appear purposive. I conclude that as computing becomes pervasive, methods in which design is considered as a form of 'reflective practice' may need to be adopted in systems design.