ESEC/FSE 2017

Gregor Engels
Paderborn University,
Germany

David Shepherd
ABB Corporate Research,
USA

Industry tracks have gained momentum recently, raising an ongoing debate about the practical value of research? From your point of view, what are the largest gaps to be bridged between foundational research and its application in practice or industrial research, respectively?

Answer: One of the fundamental problems still facing us is that we penalize academics who choose to engage in tech transfer activities. For instance, I’ve never heard of a tenure case that was helped by pointing to the number of users of a professor’s software or the number of companies who’d adopted his/her work. On the contrary, I’ve talked to professors who bemoaned the lack of credit they get for their more practical contributions, even professors whose work is used daily within advanced groups of major companies such as Google and Microsoft.

To pick up upon the first question, what would you think or suggest how we can overcome these gaps? Any countermeasures or advices for researches, especially those not having immense industry contacts? How can I do a reality check on my own?

Answer: We *need* to reward academics that engage in tech transfer. I’m not arguing that we should not still have academics that engage in completely fundamental research—we should!—I’m arguing that we should *also* have metrics that show the impact of tech transfer activities, such as number of downloads or number of active users. Until a professor with a “weak” publication record but a strong set of tech transfer metrics gets significant funding and achieves tenure we cannot expect our community to have significant impact on the practice.

Meanwhile a couple of conferences provide an industry track. Is there anything that you would say is distinct to the ESEC/FSE industry track compared to other, similar tracks (e.g., regarding evaluation criteria, types of paper, etc.)?

Answer: No. Every industry track is, and should be, the same, featuring the application of research from that conference in a realistic context. ESEC/FSE’s differs only in that it focuses on the transfer of research typically presented at ESEC/FSE.

Although industry track papers not necessarily provide new approaches or techniques, 4 pages are yet a challenging page limit. Why did you decide on such a short page limit?

Answer: We have actually taken this feedback from yourself and others and have extended the available pages to either 4 OR 6. We want to allow practitioners to submit to our track without requiring them to spend weeks polishing a 10 page paper, so we tend to err on the side of short papers.

Although I know that we as computer scientists hate magic numbers: what would be a pretty cool number of submissions you would like to have?

Answer: 50. A healthy conference should attract about 25% of the number of papers submitted to the research track, which I believe is usually around 200 at ESEC/FSE.