Benchmarking has been used to compare the performance of a variety of technologies, including computer systems, information retrieval systems, and database management systems. In these and other research areas, benchmarking has caused the science to leap forward. Until now, research disciplines have enjoyed these benefits without a good understanding of how they were achieved. In this talk, I present a process model and a theory of benchmarking to account for these effects. These were developed by examining case histories of existing benchmarks and my own experience with community-wide tool evaluations in software reverse engineering. According to the theory, the tight relationship between a benchmark and the scientific paradigm of a discipline is responsible for the leap forward. A benchmark operationalizes a scientific paradigm; it takes an abstract concept and turns it into a guide for research. Application of this theory will be illustrated using an example from reverse engineering: the C++ Extractor Test Suite (CppETS), a benchmark for comparing fact extractors for the C++ programming language. This talk will conclude with a discussion of how insights from studying benchmarking can improve the science in software engineering and collaboration in scientific communities more broadly.

(No replay available due to a shortage of technical staff to record event on the day)

Watch the webcast replay >>