Paterno `, will help to make that less likely! Many important themes are raised in Diaper (2002) review and in the commentaries. In this response I want to concentrate on refining a few of the main issues and distinctions. First, I briefly consider contrasts in values. I think that preferences for scenario-based design versus traditional task analytic methods reflect differences in the underlying world views and values of designers and researchers. To some extent, therefore, the issues in this discussion are not just technical, but paradigmatic and moral. Second, I consider the state of design practices in HCI and its implications for where onus lies in contemporary methodological argumentation and for what sorts of data are pertinent and required to address outstanding methodological issues. Finally, I consider the question of how task analysis, in the sense of structured task decomposition, might be reconstructed in the emerging paradigm of scenario-based design.
Design is about values
Design is a multifaceted endeavor that always raises fundamental questions of method and values. One must always be skeptical of claims that a design method 'works', because in design, works is a loaded term. It is loaded with values. For example, a designer cannot simply mouth HCI motherhoods like 'the design should optimize human performance', 'the design should afford immediate learning by exploration', 'the designed system should