ABSTRACT
Additionally, the task these designers have been assigned lends itself well to this investigation, as it is essentially a so ware modeling task for a real-world phenomenontrac. As such, discussion content in each of these sessions explicitly concerns itself with attempts to determine which aspects of the world (e.g., time, cars, roads, lanes, signals, distances, speeds, trac density, etc.) need to be included in the so ware (and how) for it to do what it is supposed to do: model trac ow. In any case, assumptions gure explicitly in each of the three sessions. Initially, then, it was fairly straightforward to nd some cases where designers specically deploy an “assumption” in the course of what they are doing.
