I have mentioned quite a few times in various discussions (and in a previous post) that I don't see a bright future for Debian. For some time now I've felt that Debian doesn't serve me as a distribution.

Today, this feeling got even stronger when I run in to certain issues with my server. Actually, the spark for this post was caused by this post in the Debian developers mailing list. The post reminded me about the fact that other distributions work really hard to please Debian and the developers behind it.

All in all, the situation is kind of reversed. If Debian wants to hold on to its position as the distribution this needs to change. Debian should be the one that provides these kinds of tools. If you look at Ubuntu and the tools that have been provided by canonical like launchpad (even if it is proprietary) and Ubuntu's Debian patches repository it becomes quite obvious that we really need tools to allow better integration to derivate distributions. The Utnubu team is doing a good job, but we still need tighter integration with the derivates.

Ubuntu is a good example to pick on because it's big. In some areas it's getting bigger than Debian itself. In a way, this is a good thing, but the developers should start looking at it as such. The end users will never understand the relation of derivates, but bringing more people to use the platforms known as Linux or Debian in itself is a good thing. On other areas it's not such a good thing, I personally see Ubuntu as a desktop distribution, but better integration would help to bring the features from Ubuntu to Debian that matter for the core distribution and allow the server side to be taken care of some other derivate or by Debian itself.

This brings me back to my original topic, my server. I'd love to use Debian on servers, but with the current state of the distribution I don't think it's worth it. Debian etch has been frozen for way too long. I understand that the distribution needs to be close to perfect at the time of release, but in the current state that will never happen. This in turn will cause way too long release cycles and problems like the glibc versions. To me, this means that I won't be having the latest software with as many bug fixes (upstream) as possible.

Remember, I'm not running a business class server here. It's just a home server, I couldn't care less for business grade fine tuning, as long as it runs I'm happy.

What I'm noticing from my own behaviour is that I'm starting to steer away from Debian distributions on business servers too. This is because of the problems I'm having at home. If I know that something is causing problems on small scale I'm not going to give it a shot in a large scale. This has nothing to do with the original point, but it is interesting to notice.

Comments on this page are closed.