Communicating conceptual integrity in distributed systems through intelligent assistance

The importance of ‘perceived usefulness’ as a key determinant of software acceptance has recently been widely tested. However, little attention has been given as to how the ‘perceived usefulness’ of a piece of software may be communicated to the end user through system design. The difficulties of such communication are exacerbated in distributed systems where design is typically complex and frequently undertaken by a large number of people. This paper provides a set of principles for an intelligent assistant which facilitates the effective use of distributed systems. It describes three major architectural components of an intelligent assistant, namely, an articulated help base, a user model and a dialogue control module.