It is now becoming a yearly tradition, like celebrating new year with friends and going to the music festivals in summer. Every year in february fever is rising when activity on the mailing list refreshens my mind that it is nearly that time again. FOSDEM. Time to see your Open Source Heros in real life, time to see some developers of your favourite free software projects, time to get up to date with some free software products that you didn't dream would exist, time to find out first hand about the future of some of the products you have been using. In short, exciting times are coming.

This year's hero talkers were the leader of the wikipedia project, Jimmy Wales, and the maintainer of the kernel tree at Red Hat, Alan Cox. Not that I'm a big fan of either, I don't use Red Hat and I have only recently discovered the merits of a thing called Wikipedia, but these guys are so well respected that that does not matter anymore, I just go and listen to what they have to say.

For the opening talk about Wikipedia I was only just in time, having the annual trouble of finding my way around in Brussels. Somehow I seem to be capable of always finding different directions to get lost in, which is a bit worrying really. Jimmy Wales was here basically to find volunteer developers to help on the Wikipedia project. He was also conducting some studies trying to prove that there is a big community of writers in Wikipedia, but only a small minority of them is a regular reviewer of articles. That's interesting, but it seemed to me he was worried about the criticism of classic encyclopedias that there are no expert reviewers, and was therefore trying to prove there was a small minority of volunteer reviewers which could then be called Wikipedia experts. To me, it is not interesting whether or not Wikipedia is better than a Britannica encyclopedia, and it would seem like something I would not worry about. Does it matter if the text is a 100% accurate? Not really, I think. It is up to the reader of any source to take the source for what it's worth and it helps to have different sources and compare them to find out one's own view on a particular topic. The more sources, the better, and therefore Wikipedia is an enormously valuable resource. But, as he was proudly explaining, the amount of hits he was serving on wikipedia was twice as big as an American newspaper was having and he only had one part time employee working while the American newspaper was having a hundred employees. Competitiveness which is, in my humble opinion, irrelevant. Victory is imminent, why should it be fast? It has long surprised me that universities don't have public knowledge websites, or don't even try to have one. That renouned scientifical magazines publishing articles and ideas are so expensive they effectively form a closed hard to get in community. It's all about the money of course. And Wikipedia sure can't compete yet, but see where they've come from: in not even ten years, the community can nearly compete with encyclopedias with hundreds of years experience, and these encyclopedias are getting worried.

That's a general finding when walking around and seeing various talks here at FOSDEM, and somebody else already made the remark in the forum: it is amazing to see the enthousiasm here that developers feel for their product. And it's even more amazing if you're coming from a business environment where the enthousiasm is directly proportional to the amount of money a certain technology or product is expected to deliver. Is it any wonder that refreshing ideas are to be found in this community?

The first real tech talk I went to is the kdevelop talk. The main thing I got from it is that there is going to be a Mac version and that's good, because I'm not really impressed with any of the development environments I've seen on Mac and I'm sure it's possible to do better. When the Mac version is there I am really going to check it out.

Being a Mac guy, and having only recently followed a cocoa course, I was quite enthousiastic about the occurence of a gnustep room, which is the open source counterpart of the Cocoa framework Apple gives. These developer rooms, lightning talks as they call them at fosdem are really a good idea: there are a lot more options to choose from and as a result there are fewer people, and mostly the people that are really interested. The talk was an introduction talk, so there weren't as many new things as I would have liked to have, but that's what you have with introduction talks I guess. There was more to come, but since I promised myself to follow the talk about Scribus, I had to go away. That's the disadvantage of having to choose.

Since there were a lot of changes of rooms in the tracks I had some problem finding the talk, but I did in the end. This Scribus guy however hadn't prepared his talk too well and therefore sitting there while something more interesting might have been going on in another room felt like a waste of time sometimes. However, some interesting questions came out of the public and in the end a discussion between him and the KDE guys even sounded like it was going somewhere.

I wanted to arrive on time the next morning, but that didn't work out really since the breakfast at the hotel looked a bit like it was a waste to eat in a hurry, so we took our time there. Which meant I was late for the Camino talk, and only heard the Q&A sessions there. It did not look like I had missed much and one of the most important things I wanted to know about Camino was answered: no Mozilla Extensions in the near future. That'll probably mean that I'll just stick with my full Mozilla suite, which is still my favourite browser. Sitting here remembers me of one of the first times around at FOSDEM, when a guy stood up and asked the Mozilla people the question "When will I see a workable browser?". A discussion resulted and basically it came down to the fact that Mozilla people were admitting that coding a browser for them was all about fun. They did not really hurry with their development and in pure marketing terms they were fighting a lost battle, since those were the days that everybody was using Internet Explorer and there was hardly an alternative. Look where we are now. The same guys are still there, and they are as enthousiastic now as they were then. XUL was then a new feature and XULrunner was the new feature of today, but unfortunately the demo went a bit amiss because of problems with X86Config and the monitor and the laptop or something like that. But, there was this old fashioned blackboard and with some help of a good memory we got some interesting info about it anyway. I should really try and build a little XUL-app someday, to see what it's like, but I know I probably will never get there due to a lack of time. Good luck to the Mozilla Folks, may they be forever independent and forever enthousiastic!

Everybody was getting ready for the guru talk you couldn't miss: Alan Cox was showing up, to talk. It turned out he was going to explain the philosophy of his kernel tree versus Linus' tree, with sound and practical reasons. He succeeded, I think, or at least his philosophy made sense to me. There were some good quotes to keep the attention of the audience too. Like "Linus is a good developer but a bad engineer". Now, I can't talk for Alan, but to me it would seem like he used bold talk to get his point across that there are merely different interests in Linus' tree than in Alan's, but of course those are the things ZDNet are likely to pick up too. Let's get some controversy going. There was criticism to Linux, on the security point, however. Alan proposed a solution to have a small set of people in a critical security mailing list instead of the current practice of Linus to assume that if you don't mark the patch as "security fix" the bad guys won't be able to tell the difference between that patch and the next one. That's constructive criticism, isn't it?

There was a bit of a period after Alan's talk where I couldn't find anything really interesting on the program, but I did not feel like going home before the closing talk about GPL Enforcement. So I just went to the HA talk from a totally neurotic but terribly funny guy. HA stands for High Availability and it was an interesting talk but it just wasn't a topic I am interested in. Surfing on the slow but working wireless network I found I had missed interesting bits of the Dokeos track, which would have been more my bag. Oh well. Picked up a talk about gentoo on the powerpc, but did not really learn much from that either. Another thing I should check out, right.

So, there came the closing talk. I had seen this guy on fosdem before, he was explaining about iptables internals the other year. And, I remember it vividly, he was pretty blunt about his view on software patents in those days, saying that if software patents were ever to be accepted in his country he would move to Brasil. Interesting guy, good talker. His development is mainly sponsored by astaro linux and of course iptables is very interesting for vendors of router products to embed, so therefore I can perfectly imagine he's got quite a job tracking down all devices with his code in it. Since it's released under the GPL, companies using his code are legally required to give back any changes they've made before embedding the product, which I can imagine some companies like to "forget". Name and shame and sue them was what he was doing, and he was here to give some practical insights on the issues he encountered and on the findings he got from talking to various lawyers. Really interesting stuff and I am glad to have sticked around for this!

It's fantastic to have such an interesting event happening every year in my backyard. Of course I will attend next year! No doubt about that.