This article proposes a research agenda aimed at enabling optimized testing and analysis processes and tools to support component-based software development communities. We hypothesize that de facto communities--sets of projects that provide, maintain and integrate many shared infrastructure components--are commonplace. Currently, community members, often unknown to each other, tend to work in isolation, duplicating work, failing to learn from each other's effort, and missing opportunities to efficiently improve the common infrastructure. We further hypothesize that as software integration continues to become the predominant mode of software development, there will be increasing value in tools and techniques that empower these communities to coordinate and optimize their development efforts, and to generate and broadly share information. Such tools and techniques will greatly improve the robustness, quality and usability of the common infrastructure which, in turn, will greatly r...
Atif M. Memon, Adam A. Porter, Alan Sussman