I’ve just returned from ICFP 2009, the The 14th ACM SIGPLAN International Conference on Functional Programming in Edinburgh.
The conference went well, and was even more interesting than last year, perhaps because I comprehended a lot more of it. Last year seemed to be the year of dependent types, about which I had little clue at the time; this year I seemed to be hearing all the time about applicative functors. (Not that people were giving papers directly about applicative functors, but there seemed to be quite a number that chugged along in their own direction for a while and suddenly mentioned that what they were working on just happened to be an applicative functor as well.)
There was more Haskell stuff than ever, with not only the Haskell Symposium but an additional day added for the Haskell Implementors Workshop. Particularly demonstrating the interest in Haskell was a quick quiz at CUFP where the discussion moderator asked how many were currently working on non-research applications in various languages, and when he asked about Haskell it seemed that eighty percent of the room put up their hands.
I did two presentations, one of my paper “Haskell in the Real World” at ICFP itself, which was quite well received, and a more informal presentation on how I develop software at the Haskell Implementors Workshop, in the context of Cabal and other build systems. The latter was interesting to me as I found more interest than I’d expected in some of the mechanics of reliably going from a bunch of code to a released product. Particularly surprising to me (but also quite gratifying) was a comment from Norman Ramsey that he had similar issues even in the academic world where he relied on a lot of other software he’d not written, and came up with these “Frankenstein’s monster applications” that were difficult for others to set up and run.
The results of the programming contest were announced, and I’m very pleased to say that dysfukycom, a team consisting of the Lisp variety of TSAC members, placed in the top ten.
In other good news, while ICFP 2010 is in Baltimore, it’s been confirmed that ICFP 2011 will be in Tokyo. Not only am I very much looking forward to attending the conference without having to bookend it with a twelve-hour plane flight at either end, but it will give TSAC members a chance to attend some of the sessions at a fairly reasonable price. (ICFP itself is fairly expensive, but CUFP, the Haskell Symposum, and so on generally run $100-$200 each.)
I took notes throughout (close to a thousand lines of them) and if I can find the time I’ll try to put up a summary of the presentations, as well as my slides from both of my presentations.
Which does bring up one annoyance: the electronic version of the proceedings is still not available from the ACM. I have a paper copy, but it’s annoying to lug around between home and the office, and of course it means I’ve got to photocopy articles I’d like to share and discuss with my friends in the programming community. I hope that the electronic versions of the papers become available soon.