Anthony Towns <aj@azure.humbug.org.au> writes: > problems before you upload to unstable. The differences are, eg, between > making sure Gnome 2.6 can be built everywhere, and actually having a > particular version of Gnome 2.6 built everywhere;
Gnome is usually really good on the portability plan (no problem with 2.2, 2.4), there is no reason to fear massive problems for gnome2.6. It has built on 7 architectures without problems and I don't think we are magically getting a lot problems of the few missing. And if we get some problems, we are quite a lot to work on gnome packages and I really think we are able to deal fast with these problems ... > between fixing RC bugs > you know about in advance, and spending a couple of weeks letting users > find new ones and fixing those too. We are fixing bug for 2 months now and you can check on the BTS to see gnome2.6 bugs reported and how we are dealing we them. According to #gnome-debian, gtk-gnome list and BTS activities there are a lot of users who have already tested these packages. > If they're really that good, there's no big deal at all. This is how > you convince us they're really that good. Packages are in experimental for 2 month. Widely tested, fixed, built, ... look on the BTS, gnome list activity, .. And the usual good quality of gnome releases is also a good sign ... > pretty majorly: packages in unstable with RC bugs need to be fixed > _quickly_. Not immediately, maybe, but not after months and months > either. Once again we are very reactive (look on RC bug reported on gnome2.6 packages in the BTS for the 2 last months) > habit of risking. The way we can avoid it is by running major changes > through experimental first; these things won't catch everything by any > means but they will catch a bunch of issues that do cause major problems > elsewhere. Are you asking for wide use of experimental ? Now unstable is supposed to play testing role and experimental unstable one ? Manual builds in experimental give a lot of extra work to maintainer, it's not possible to keep this effort continualy, that's wasting too much ressource for no gain ... Cheers, Sebastien Bacher
