Supporting reuse by delivering task-relevant and personalized information

Abstract

Technical, cognitive and social factors inhibit the widespread success of systematic software reuse. Our research is primarily concerned with the cognitive and social challenges faced by software developers: how to motivate them to reuse software, and how to reduce the difficulty of locating components from a large reuse repository. Our research has explored a new interaction style between software developers and reuse repository systems enabled by information delivery mechanisms. Instead of passively waiting for software developers to explore the reuse repository with explicit queries, information delivery autonomously locates and presents components by using the developers' partially written programs as implicit queries. We have designed, implemented, and evaluated a system called CodeBroker, which illustrates different techniques to address the essential challenges in information delivery: to make the delivered information relevant to the task at hand and personalized to the background knowledge of an individual developer. Empirical evaluations of CodeBroker show that information delivery is effective in promoting reuse.