Software reuse 'in the large': Tracing patterns of reuse

While there has been much research on software reuse at the fine- and medium-grained levels, there is little published on actual case studies of very large-grained software reuse within corporate environments. This study focuses on BigFinancial, a large financial services corporation with significant software development capabilities, and their experience with reuse at fine-, medium-, and large-grained levels. Factors contributing to successful reuse include organization, technical, cultural, and infrastructural elements, as well as the track record of the organization in successfully reusing systems. The study extends the notion of "large-grained" reuse to include the reuse of entire application systems and all underlying layers, across separate and independent business units. The study looks at the business value of reusability relative to the alternatives considered by BigFinancial, as well as the organizational challenges faced in securing agreement for such deployments.; The study extends the literature in the area of large-grained software reuse. Fine-grained reuse literature provided a foundation on which to build, and much of the theoretical foundation for fine-grained reuse success was supported in this study. Notably, though, neither incentives nor reuse metrics or reuse maturity were significant contributors to the success of the reuse program.; Several new causal factors were identified in the course of the study. While there is limited mention in the literature regarding trust and culture, the study shows that trust in the development group is a key contributor to the success of large-grained software reuse projects. In addition, the organizational culture of both the development and business user organizations must be receptive to large-grained software reuse. Further, corporate mandates were found to be a key factor in at least some of the reuse projects.; Finally, a very succinct measure of reuse success was identified. BigFinancial defines success via the simple measures of the number of separate installations of a particular product, and the number of ways that product is being used across the potential user base. While this success model is most applicable to large-grained reuse, it may also have utility for finer-grained reuse success measurements.