Heuristics for use case descriptions.

Abstract

Use cases, as part of the Unified Modelling Language, have become an industry standard. The major focus has been on the use case diagram. It is only recently that any detailed attention has been paid to the use case description. The description should be written in such a way as to make it communicable to its reader. However, this does not always appear to be the case. This thesis presents the 7 C's of Communicability as quality features of use case descriptions that make them more comprehensible. The 7 C's are derived from software engineering best practice on use case descriptions and from theories of text comprehension. To help in writing descriptions, the CP Use Case Writing Rules are proposed, a small set of guidelines derived from the 7 C's. Going beyond requirements, software engineers often employ use case descriptions to help them build initial design models of the proposed system. Despite Jacobson's claim that "objects naturally fall out of use cases", fording design-oriented classes and objects in use case descriptions is shown not to be straightforward. This thesis proposes a Question Set which allows the engineer to interrogate the description for important elements of specification and design. Experimentation shows that the CP Writing Rules furnish descriptions that are as comprehensible as those written by other guidelines proposed in the literature. It is also suggested that descriptions be written from the perspective of their intended audience. The limitations of conducting requirements engineering experiments using students are considered and it is suggested that experimenters should not expect large effects from the results. An industrial case study shows that although the CP Rules could not be applied to all events in the use case descriptions, they were applied to most and at varying levels of abstraction. The case study showed that the 7 C's did identify problems with the written descriptions. The Question Set was well received by the case study stakeholders, but it was considered time consuming. One of the overriding findings from the case study was that project time constraints would not allow the company to use the techniques suggested, although they recognised the need to do so. Automation would make industrial application of the CP Rules and 7 C's more feasible.