Specifying use case interaction: Clarifying extension points and rejoin points

Abstract

This article is the second work in a series of papers on advancing specification and modelling techniques for goal-based use cases. In this article we highlight major and remaining problems associated with the true semantics of extension points and rejoin points, which are used as branching and return locations for a use case's alternative interaction courses, and also for the application of UML's Extend-relationship. Based on a previous work that identified and defined three reconciling types of alternative courses of use case interaction, this article proposes clear and simple semantics for extension points and rejoin points that significantly reduce use case interaction ambiguity. These semantics allow the uniform treatment of any type of alternative course, so that a practitioner can concentrate on the coherence of the use case's goal, precondition and business results. Our proposal is independent of the way the use case interaction is documented, i.e. textually or graphically. Finally, our suggested semantics will dramatically contribute to an understanding of practical use case interaction specification.