Back from my first ICFP and CUFP, where I had my first talk in the functional languages domain. Here is a link to the recorded talks, mine was “Haskell in the Mission Control Domain”:
It was a very good experience and also quite impressive, to see all the creme-de-la-creme from the Haskell community in one place! Also it was nice to get an association from names to actual faces.
Ok, so a bit more details (from what I remember):
CUFP day 1: The “Real World Haskell” Tutorial bei Stefan Wehr. Since this is also a region which I work most, this was a must for me and I really got a lot of new ideas on some things (esp. logging and tracing, which in my opinion is still lacking in Haskell), and also how other people approach applications
In the afternoon I was at Edward Kmett‘s Lenses Tutorial, which was very interesting but way above my head. I was very impressed how fluid he is in all that stuff and directly “live-codes” this abstract high-level type stuff. I decided that I have to take a deeper look into category theory.
In the evening there was the reception in the Museum where all the industrial partners introduced themselves (and also presented their job offers). Wine and tappas were available and I really had some nice talks with some of the people there.
CUFP day 2: Rust Tutorial: it is a quite nice language and I see it’s uses, but it needs to mature a lot more before it is ready for production use. Also it has to compete with languages like e.g. Go and D on the more imperative side, so we will see, if the more functional approach will bring benefits. Anyway, there were also a lot of interesting questions from people in the tutorial, which also showed what the people are concerned with.
In the afternoon I was at the QuickCheck tutorial from John Hughes and that was really enlightening. He showed on how to find quite difficult bugs directly via live coding in Erlang and how they use it at Quvic. This also made me think about a lot of the idiosnycracies in SCOS and if that could also help to find them… In a break I had the chance to talk with John and on their approach to test API’s (mostly C) and how this could be applied also to C++ code.
Saturday: on Saturday the CUFP talks were held, inclusive mine and Björns. Joe Armstrong really gave a fabulous talk on how to earn money with functional programming. There were a lot of tips inside his talk, and also he is a quite funny guy. I also was at the Haskell Implementers Workshop, because, of course you can’t miss a talk from Simon Peyton Jones. I really love them. The room was very crowded, so the interest is really high.
Joachim Breitner also did a live demo on how to contribute to GHC development. Could be interesting, but I have massive time problems currently, since I am doing a lot more than programming.
Also the last CUFP talk from Andrew Cowie, “If your server is a function, is your company a library?” I enjoyed a lot (it is also present in the playlist above). It was about an approach of testing larger distributed systems, which can be quite challenging as I know from my own experience. They started to rewirte the server software as Haskell libraries, so finally, they could link everything together and test it as one piece locally. Since he was the last talker, there were also not so much people anymore, but his talk was very interesting. I also talked with Andrew afterwards which was very nice.
Food was quite good, I really liked my hotel (Scandic Opalen Hotel), organization was also quite good (they also had services to print your boarding card, which I really appreciate).
On my flight back, I found a really lovely bar in the Landvetter airport called Joe & The Juice where you get juices from organic fruits and also nice sandwiches. They also have really comfortable seats, so I was able to chill down, despite the early time (it was 5:45 in the morning). At the gate I met Michael Snoyman and two other guys where I forgot the name, and we also had a nice chat.
All in all, it was a really nice conference, the Haskell community is really really friendly, which is also very interesting from a psychological viewpoint (if you view it from e.g. the Graves Model or Spiral Dynamics).
So all in all, it was good experience.