Software Reuse in Information Systems Development

Software and Management Information Systems application development have become a key area to the performance of most firms. The reuse of previously written code is a way to increase software development productivity as well as the quality of the software (Basili, et al., 1996; Gaffney and Durek, 1989). If previously tested components are reused in a new software project, they are more likely to be error free than new components. This reduces the overall failure rate of the software project. Case studies, such as (Banker and Kauffman, 1991; Poulin, et al., 1993; Apte, et al., 1990; Lim, 1994; Swanson, et al., 1991) were instrumental in obtaining such insights. Today, an increasing number of organizations are adopting the practice of software reuse (Lim, 1994). A common misconception is that object-orientation alone will lead to reuse. While it can help facilitating a reuse approach, research has shown that object technology does not always lead to reuse (Fichman and Kemerer, 1997). Software reuse requires a substantial up-front investment for the development and maintenance of a software repository with reusable components (Barnes and Bollinger, 1991). A large part of the set-up cost comes from the fact that additional effort is needed to make regular components generic enough for use in future projects (Mili, et al., 1994). In the long-run, this initial investment can be offset by the cost savings through reuse. Using a reusable component in lieu of writing a new component from scratch saves development cost. However, the component has to be located and retrieved from the repository. Often, components cannot be used as is, but also need to be modified to fit the context of the new project. Reuse can only be economically viable, if the savings achieved through reuse will over time offset the start-up cost of implementing the reuse methodology and populating the software repository. While software reuse is not a new research area to computer science (Krueger, 1992), MIS research has only recently begun to investigate this important aspect of software development. This reflects an increased understanding that too little work has been done on nontechnical issues (Zand and Samadzadeh, 1995). Organizational and behavioral aspects, legal constraints, and economic considerations are little explored in the context of software reuse. IS research can also contribute to storage and retrieval problem by developing domain specific solutions. In this overview we summarize the work done in the areas most important to IS research