On Saturday 09 June 2012, Albert Astals Cid wrote: > There is not stupid not weird thing in kdelibs. We simply declared kdelibs > to be feature perfect and only bugfixes should be happening there. > > What's the problem with that?
That no software is ever perfect. For example, we will NEED the Solid udisks2 backend in future releases of Fedora because udisks 1 is going away! As far as I know (I'm not a Red Hat employee), it will be used in the next release of RHEL as well. So we have to backport it as a distro patch. (The author of the backend did the backport.) For the other distros: http://pkgs.fedoraproject.org/gitweb/?p=kdelibs.git;a=blob;f=kdelibs-udisks2_prep.patch;h=2962837cdbc3cc1200ce323902f9173b79c2b1b5;hb=HEAD http://pkgs.fedoraproject.org/gitweb/?p=kdelibs.git;a=blob;f=kdelibs-udisks2-backend.patch;h=74ec3c2872d118d0335cc95c608a95ef9cd40f92;hb=HEAD http://pkgs.fedoraproject.org/gitweb/?p=kdelibs.git;a=blob;f=kdelibs-udisks2_post.patch;h=f22742dc8db8b2dd2b48de6f916c4a50a360db6b;hb=HEAD kdelibs has now been feature-frozen for a year, and it looks like it will live at least 2 years in feature-frozen state, plus several more as a compatibility library. (We still ship a kdelibs3 compatibility library in Fedora, it's still needed, e.g. for Quanta.) This is just unacceptable. The libraries should be the LAST thing to be frozen, even after the core applications have already moved on, because of the many third-party applications relying on them. Freezing the libraries first, when even the core applications still rely on the "old" version, just makes no sense whatsoever. Kevin Kofler
