Tasks = data + action + context: Automated task assistance through data-oriented analysis

Abstract

Human activity unfolds partly through planning and learnt sequences of actions, and partly through reaction to the physical objects and digital data in the environment. This paper describes various techniques related to automatic task assistance that take this role of data as central. Although this brings additional complexity, it also offers ways to simplify or bypass problems in task inference that otherwise appear difficult or impossible. Although the focus in this paper is on automated task support, the importance of objects and data in understanding tasks is one that applies to other forms of task analysis in the design process. © 2008 Springer Berlin Heidelberg.