time-based realease, uh?
Hi.
Once again, Debian is hitting the news, showing to everybody that a time-based
realease plan for Debian just can't work [1]. Even its own developpers seem to
have no idea of when it will be released, which looks, I'm sorry, quite funny
[2]. What's worse is that this time it provoked a clash in the developper's
community, which was the worst thing that could possibly happen [3, 4, 5, 6, 7].
Don't get me wrong, I love Debian. So maybe it's time to to adopt a logical
approach, huh? Saying, almost 2 years before, we'll release Sarge+1 at this
precise date is not only irrealistic, but also stupid. You end up tagging many
bugs as etch-ignore, having to hurry to include packages like mplayer, postpone
burning issues like non-free firmwares, trying to find the least buggy kernel
version to include, etc.
What would be more logical, at least, would be to have a deadline for the
freeze, but not the final release. Or better, once you have the right kernel,
the good toolchain, and solved critical issues (non-free firmwares, anybody?),
then you freeze. And then, you release when all rc-bugs are solved.
Debian is all about quality and freedom, at least for me. So if you hurry to
release on time, you lose quality. And you end up having non-free fimwares and
several other issues. So you don't have freedom.
What's left?
[1] http://news.zdnet.co.uk/software/0,1000000121,39285819,00.htm
[2] http://news.zdnet.co.uk/software/0,1000000121,39285649,00.htm?r=1
[3] http://www.dunc-tank.org/
[4] http://lists.debian.org/debian-vote/2006/09/msg00267.html
[5] http://dunc-bank.zoy.org/
[6] http://www.computerworld.com.au/index.php/id;189708751;fp;4194304;fpid;1
[7] http://www.linux-watch.com/news/NS3128387759.html
Reply to: