Debian etch freeze to be delayed
From: | Andreas Barth <aba-AT-not.so.argh.org> | |
To: | debian-devel-announce-AT-lists.debian.org | |
Subject: | delay of the full etch freeze | |
Date: | Wed, 11 Oct 2006 18:58:55 +0200 |
Hi, In spite of much hard work by our many bugsquashers, the current RC bug count is floating around 250. This is much higher than the number asked for by the release schedule at the point of the full freeze. We want to avoid putting ourselves in the position of spending a lot of time reviewing packages instead of being able to continue working on fixing bugs, which is why we've set our target at 80 RC bugs. Freezing with more bugs would likely make the total time to release longer than it would be otherwise, as a result of the time spent managing the freeze instead of working on bugs. While this might prove necessary if we see too many new RC issues introduced in packages, we hope based on current bug trends that more time before the freeze will let us bring the RC bug count down to the desired level. In addition, we're a bit behind schedule for the d-i release candidate, which we want to have released and collect feedback from before the freeze. For these reasons, we are delaying the full archive freeze for a few days. We haven't chosen a date yet, but you can still expect it to happen in October or early November. In the meantime, please carry on as usual -- as long as "as usual" means "fixing more RC bugs than you open", so that we can continue to aim for a release at the beginning of December 2006. :) Cheers, Andi -- http://home.arcor.de/andreas-barth/
Posted Oct 11, 2006 19:26 UTC (Wed)
by dcg (subscriber, #9198)
[Link] (5 responses)
Some of the "release critical bugs" are bugs that affect packages like..."amaya"
Why bugs in a browser that almost nobody uses like amaya should have right to be considered "release critical" at the same level of priority than gnome-session? There're like 15.000 packages in debian. It's not surprising that there're always a huge number of RC bugs in debian: the huge number of packages implies that it's just not possible to release a RC-free release. Worse, RC bugs in more important packages like X.org/gnome/kde/apache may get less attention because of this. Sure, bugs in the base-system get more attention, but the base-system is a unrealistic view of the real "base-systems" that are used in real world installs.
It's just not fair and not productive to consider that all packages have the same priority. It happens the same WRT security issues. Debian should have some priorities.
Posted Oct 11, 2006 19:54 UTC (Wed)
by cortana (subscriber, #24596)
[Link] (1 responses)
Well... interestingly, Amaya is not presently in testing. It is not holding up the release. As with many debian.org facilities, the most useful web pages are the unofficial ones. See Unofficial RC-Bugs Count, for example. FYI, amaya is the 4338th (out of 23984) used package in Debian, with just 31 users. There is also a graph of this data. So the RMs are unlikely to hold up the release to allow it to propogate back in.
Posted Oct 11, 2006 22:17 UTC (Wed)
by ballombe (subscriber, #9523)
[Link]
Posted Oct 11, 2006 22:52 UTC (Wed)
by bignose (subscriber, #40)
[Link]
Certainly it is possible. You may be misunderstanding what it means to say that a bug is "release-critical": it is not that the *release* is in jeopardy of releasing, but that the *package* is in jeapordy of not being included with the release.
If a bug on a package is considered release-critical, one of the ways of resolving that status is to omit that package from the release. This is the last resort in cases where the bug cannot be closed (or, in a few cases, determined no longer to be high severity) in time for the release.
> It's just not fair and not productive to consider that all packages have
If Debian did treat all packages the same, that sounds like the *definition* of fair treatment to me :-)
As it happens, though, Debian does prioritise packages: http://www.debian.org/doc/FAQ/ch-pkg_basics.en.html#s-pri...
It is these priorities that lead to different attitudes when the time comes to decide on the omission of a package from a release.
Posted Oct 12, 2006 5:18 UTC (Thu)
by thedevil (guest, #32913)
[Link] (1 responses)
OK, but amaya is not primarily a browser. It can be used as one, but more importantly it is a webpage composer tool. I suggest that all the registered users (reported in the other subthread) installed it for that purpose. When that is considered it may appear a bit more important (there's no Firefox for webpage composition).
Posted Oct 12, 2006 10:18 UTC (Thu)
by lacostej (guest, #2760)
[Link]
See nvu.com and now composer (rewritten for latest gecko)
http://www.glazman.org/weblog/
(Based on old Mozilla composer)
Just take a look at: http://bugs.debian.org/release-critical/debian/main.htmlAgain, the needs of a reorganization of the whole project
Again, the needs of a reorganization of the whole project
You link to the wrong listing: recent show the number of people who upgraded this package recently.
Actually, amaya is the 3280th most used package. Not that made much of a difference...
Again, the needs of a reorganization of the whole project
> the huge number of packages implies that it's just not possible toDebian packages are prioritised
> release a RC-free release.
> the same priority.
> Why bugs in a browser that almost nobody uses like amayaIn defense of amaya
> there's no Firefox for webpage compositionIn defense of amaya