Three software engineering researchers meet at the reception bar of ICSE. The first one claims, with much conviction, that software systems are technical artefacts, many of them critical, and that therefore our duty, as researchers, is to ensure, through mathematical means, that systems are provably correct, safe, and secure. No other objective is a worthwhile academic endeavor or deserving intellectual respect. The second fellow points out that software systems are developed by people and that therefore social, organizational, and psychological aspects play a predominant role in successfully developing software. Developers must indeed be happy, diverse, and meet various other desirable qualities. Ignoring such human factors leads inevitably to unsatisfactory solutions. The last researcher, on the other hand, points out that automation is the key to many software engineering problems and that we should develop technologies that demonstrate scalability to large systems and high performance in automating whatever task they target, lest our novel solutions should remain unused. After a much heated debate, and some drinking, they part ways frustrated and angry, more adamant than ever about their perspective on research and ready to tweet their disappointment to the world. Can we reconcile these three viewpoints in our community in such way to increase the impact of our research? What does having impact entail as a researcher? Should researchers even seek impact?

http://www.lbriand.info