Abstract

The prevailing computer paradigms do a poor job at meeting their human users at their level of abstractions. Humans organize work and leisure in more or less well defined tasks and activities. Computers organize computing in terms of applications, files, networks, etc. This abstraction gap becomes a big problem in situations characterized by mobility, frequent interruptions, and collaboration – situations that pervasive computing is intended to support. A new paradigm, task-level or activity-based computing, has been proposed to lessen the abstraction gap and provide a better platform for pervasive computing. In this paper, we shortly describe our approach to activity-based computing and describe lessons learned so far, including discussing some of the more tricky open issues which has surfaced in our research.

Citations