Fighting? [LWN.net]

Fighting?

Posted Feb 7, 2007 5:06 UTC (Wed) by moxfyre (guest, #13847)
In reply to: Fighting? by proski
Parent article: Second batch of FOSDEM interviews

And what's wrong with trying more than one approach to the same problem?

I agree completely! People always say that "GNOME vs KDE" is the worst use of the free software community's time and "why can't they merge?" to make a "unified" desktop like Windoze.

That makes no sense to me at all. Instead of having one DE, we open source users have *TWO* fabulous desktop environments with slightly different strengths and weaknesses... the competition and collaboration and cross-pollination have improved them both immeasurably. How does having MORE GOOD CHOICES hurt anyone?

(Log in to post comments)

Fighting?

Posted Feb 7, 2007 9:36 UTC (Wed) by gravious (guest, #7662) [Link]

How does having MORE GOOD CHOICES hurt anyone?

In no way other than now people have to evaluate or put a little thought into what software they use, poor them - no more hand-holding. I dunno, First they complain about monopolies and then they complain about competition, you can't win, I tells ya!

Fighting?

Posted Feb 7, 2007 18:28 UTC (Wed) by eklitzke (subscriber, #36426) [Link]

Competition is good. But forks are almost always a bad thing.

If you look at the Gnome vs KDE thing, everything is OK now (a long time ago, it used to be a different story). We have XDG, which has actually improved the quality of both Gnome and KDE (and other DEs like XFCE), and has made applications largely DE independent. The two projects have different aims and a healthy number of developers, and as a user I benefit from this.

On the other hand forks can have a really devastating effect on a project. The best case scenario is that you get something like the XFree/X.org split where one of the projects dies really quickly, and everyone can just get back to work. But other times the forks just linger on, things become incompatible, efforts are duplicated, and you're never really sure what supports what. This is especially true for specialized projects like Compiz/Beryl -- there just aren't very many developers with the expertise to hack on these projects. Splitting the small talent pool just exacerbates the situation, especially when the goals of Compiz and Beryl are 90% the same. As far as I can tell, Beryl is just Compiz with a GUI config tool, better KDE support, and one or two extra plugins. Why was a fork necessary again?

The Beryl project seems to have more momentum behind it right now, so unless Compiz is adopted by Gnome or gets more support thrown behind it from another distributor, it looks like Compiz will quietly die off. But Metacity and KWin are great window managers already, not to mention <insert your favorite lightweight WM here>. This has led to efforts like libcm to create a compositing library that other window managers can easily take advantage. My fingers are crossed that this approach will prove successful, but in the meantime I don't think that the fragmentation in the community is leading to a lot of "cross-pollination".