Towards an understanding of the use of problem and design spaces during object-oriented system development

Abstract

The importance of, and distinction between, problem and design spaces has been recognized in studies of information systems development (ISD). With increasing acceptance of object-oriented techniques, which promise close mirroring of real-world concepts in the IS artifacts, this distinction becomes even more important. In spite of a rich literature stream dealing with the general notion of ‘design,’ several inadequacies remain in our understanding of the ISD processes—one of these is the recognition and impact of problem and design spaces. In this paper, we analyze processes followed by two developers engaged in a non-trivial development task using the object-oriented modeling techniques—with a view to understanding their use of, and explorations in, problem and design spaces. Our analysis provides evidence for a distinction between the two spaces, and interprets the developers’ engagements and behaviors to structure the two spaces with the help of object-oriented modeling techniques. Several interesting findings emerge from our analysis, including the overlapping of spaces due to the use of object-orientation, disturbing patterns such as design fixation, interplay between simulation, expansion and validation in the design space, and the varying use of modeling techniques to structure the two spaces. Our analysis, supported by prior literature, provides a grounded description of some phenomena that have, hitherto, had only intuitive or prescriptive support. Based on these, we argue for more methodological and tool support for explorations of problem space, and enhancements to current approaches, for better integration of problem and design spaces.