On Thu, 8 Oct 2009, Toshio Kuratomi wrote:
I like the daring of this plan. I think it has the potential to improve Fedora quite a lot. However, I think it has more potential to improve Fedora through finding all the points of failure when we do this than in having a rousing success. For instance, GNOME and KDE use several of the same libraries under the hood. What conflicts does this cause if we have to/have to not upgrade a library due to version conflicts? Finding those sorts of pain points and figuring ways to mitigate them in the future will be more valuable than finding out that the KDE spin is more popular than the GNOME spin or vice versa.
I wasn't concerned with popularity - just how well things would function
And to go along with this, I expect that we'll find lots of pain points if we do this. So we need to be prepared to deal with them expeditiously when they crop up (otherwise meeting our 6 month timetable will be hard) and at the same time revist any of the solutions that are contentious once we have time after the release.
the pain points are exactly why I am interested, too. We went through this process when anaconda switched away from the hdrlist and then to the various repository formats. Changing things which _should_ be interchangeable frequently points up all the ways they are not _really_ interchangeable.
-sv _______________________________________________ fedora-advisory-board mailing list fedora-advisory-board@xxxxxxxxxx http://www.redhat.com/mailman/listinfo/fedora-advisory-board
