FOSDEM 2008 impressions Day I

This years FOSDEM is over and I'm on my way home. It's past midnight and I still have a three hour train journey before I can get home to resurrect one of my home servers:( At least I have GPRS access from the train so enough complaining for now:)

Last time I went to FOSDEM was in 2005 and since then only a few things seems to have changed. The best and the worst change is that Saturday seemed to be particularily overcrowded.

The program for Saturday was as a bit thin in my eyes so I had the opertunity to experiment a bit with what talks to attend. One of the best talks were with Robert Watson from the FreeBSD Project telling about the comminuty side of the project. I guess one key aspect difference between the Gentoo and FreeBSD developer communities is that the average age of the FreeBSD developers is 31½ year. I'd be surprised if it is over 30 in Gentoo. I tend to think that you get a bit more "round" with age.

Speaking of Gentoo, this year we unfortunately didn't get a Devroom and that was a bit of a shame as I only bumped into a few other devs. I probably could have been a bit more active in that regard but I did get to meet up with a few of the current GNAP users/developers that are trying to keep the project going. Also some Gentoo devs were still giving talks with other upstream projects (at least tester, remi, betelgeuse). Unfortunately I discovered this too late to see most of the presentations. My preparation and planning could have been much beter, but I unfortunately don't have too much time for Gentoo stuff these days.

After that I listened to the enthusiastic talk by Pieter Hintjens from FFII about the status of software patents in the EU. He's a great speaker even without manuscript and used the oppertunity to present their new campaign Kill Software Patents.

The last memorable stuff from Saturday was the presentation of SELinux in Fedora and I wasn't too impressed. I've never really had the time or wish to learn SELinux but thought this would be a nice oppertunity to learn a bit more. However one minor detail of the talk puzzled me . The default enforcing policy for Apache restricts what ports it can listen on, which is all fine. But it appeared as if the default policy allowed Apache to listen on some not so standard HTTP ports like 8180. If I should write a bot I would let it try some of those accepted ports before anything random. Also configuration still seems to be a hassle unless you know exactly what the appliation is supposed to do and how.

This is all for day one.