Wed, 25 Mar 2009

GNOME 2.24 in Debian unstable, and the road ahead

GNOME 2.26 was released last week, and I couldn't help adding myself to the long list of celebrating posts in Planet GNOME. Looking at the release notes, it looks like this release adds a good number of very visible features, and also keeps improving on ongoing transitions like gvfs.

The Debian GNOME team is obviously not ignoring this fact and started to work very hard on updating GNOME for squeeze as soon as the lenny freeze was over.

First, the new versions of GLib and GTK+ were uploaded to unstable, and managed to transition to testing very easily. The rest of GNOME 2.24 bits, which had been patiently waiting on experimental for months, has been uploaded with care not to disrupt any of the many transitions the Debian release team is currently dealing with. You can have a quick glance at how things are going in our 2.24 status page, but the summary is that most of GNOME 2.24 is in unstable, with a few notable exceptions which are held back by ongoing testing transitions. Namedly, evolution-data-server is trying to trickle into testing, which is in turn holding the final bits: gnome-panel, nautilus and related packages, but we think this will be over soon.

As soon as GNOME 2.24 is safe in squeeze, we'll immediately turn our focus to the new GNOME 2.26 release. Our initial plan is to package the trivial bits and leaf packages which can't break stuff for unstable, and herd the more complex modules via experimental, to avoid breaking unstable at all. There are some exceptions; we plan to keep gnome-session 2.22 in unstable/testing until 2.26.1 is released to avoid getting a broken session saving in Debian.

People might wonder why we insist on hitting what would seem a dead horse by first dealing with 2.24 and not 2.26 directly. The main reason is that these packages had been ready for a long time, and were in good shape to transition to testing quickly and with little pain. Preparing 2.26 directly would mean throwing away a lot of hours of packaging and polishing effort, and it's not like we're releasing squeeze any time soon anyway.

Enjoy the hopefully not too bumpy road to 2.26!

00:32 | [] | # | (comments: 7)

wOOt! i can has gnomes in squeeze :D

Posted by Emilio Pozuelo Monfort at Wed Mar 25 01:13:22 2009

FWIW, the 2.24 packages from experimental have been very useful to me for the last few months; So your work has been appreciated, it wasn't a wasted effort at all from where I'm standing. Thanks.

Posted by nona at Wed Mar 25 01:33:46 2009

http://en.wikipedia.org/wiki/Sunk_cost_fallacy#Loss_aversion_and_the_sunk_cost_fallacy

Posted by TheGZeus at Wed Mar 25 05:31:51 2009

First,thanks for this update. Really appreciated.
What about Nautilus?

Posted by Tim at Wed Mar 25 08:17:24 2009

thanks. using a mixture of experimental and sid is a bit inconvenient. one question though. is it always going to look like this that debian skips one or two releases of gnome every 2 years?
if I remember correctly the same situation was with gnome 2.16. there were 2.14 and 2.18 in debian, but no 2.16.

nevertheless thanks for the news

Posted by shutdownrunner at Wed Mar 25 09:16:06 2009

@nona: there is no sunk cost in sending GNOME 2.24 to unstable and make users benefit of it immediately instead of GNOME 2.26 in two months. The fallacy here would be to think the changes we made for 2.24 are lost for 2.26. They are not, since upstream releases are an evolutionary process.

@Tim: this is a problem of how we handle releases, by freezing unstable. From what I heard of the release team, this will be handled differently for squeeze, so we might see the end of this kind of issue.

Posted by Np237 at Wed Mar 25 10:54:03 2009

First,thanks for this update. Really appreciated.
What about Nautilus?

Posted by Tim at Wed Mar 25 11:00:34 2009