Understanding Context: Creating a Lasting Impact in Experimental Software Engineering Research

摘要

Software is developed for and in a vast number of contexts. Some software systems are small in size; some large. Some systems are developed by small teams; some large. Some projects are sensitive to schedule, others to safety of the users. In this position paper, we argue that to make a lasting impact with the software engineering research we conduct, we must understand, make explicit, and vary the context in which our conclusions are drawn. Moreover, we need a better understanding of how research results can be translated or generalized to other contexts, as it is not economically feasible to replicate results across all contexts. We argue that a successful solution to this problem will allow researchers to conduct research within particular contexts, richly characterize those contexts in their writings, and allow other researchers to predictably build on those in differing contexts.