Community

I once had a provost who felt that academic departments hindered the university as they tended to silo the faculty. He would argue we should eliminate departments and that would increase cross-disciplinary work. That went nowhere of course.

He missed a critical point. While departments play an official role in hosting academic programs, they importantly serve as the main community for the faculty within the department. You see your colleagues in the department in meetings, at seminars and just walking by them in the hallways. They are your peers and the ones who hired you and judge your promotion and tenure cases. You will argue with them but you all share a common mission to make your department as strong as possible, so you can attract even better colleagues.

At Oxford, most faculty have a position at one of the 36 colleges, and faculty community comes in the college instead of the department. Many faculty walk from their department offices to have lunch in their college. The colleges are interdisciplinary, there are only at most a couple of computer scientists in each college. My visiting fellowship is with Magdalen College, not the CS department though I plan to spend some time there.

Does the Oxford system lead to more interdisciplinary research, or less research in the field? Not sure, and it's hard to run a controlled experiment.

We all have two communities, one where we work, but also our academic community, the people who do research in the same areas, the people in our field that we collaborate with and see at conferences and workshops. The ones we try to impress with our own research so they may accept our papers, give us grants, nominate us for awards and write good letters for us. That's the real challenge of interdisciplinary research--you hope to impress researchers in both fields, but really you just impress those in the intersection.

In computer science our academic communities are pretty small, usually the conferences where we publish regularly. I was having a conversation with another computer scientist at Oxford. I asked him which conference and he said ICALP B. I said I've published in ICALP A, completely different!