Evaluation of Prototypes and the Problem of Possible Futures.

There is a blind spot in HCI's evaluation methodology: we rarely consider the implications of the fact that a prototype can never be fully evaluated in a study. A prototype under study exists firmly in the present world, in the circumstances created in the study, but its real context of use is a partially unknown future state of affairs. This present-future gap is implicit in any evaluation of prototypes, be they usability tests, controlled experiments, or field trials. A carelessly designed evaluation may inadvertently evaluate the wrong futures, contexts, or user groups, thereby leading to false conclusions and expensive design failures. The essay analyses evaluation methodology from this perspective, illuminating how to mitigate the present-future gap.