Fedora 13 continuing the tradition of being an update monster
mhlavink at redhat.com
Tue May 11 11:26:47 UTC 2010
> Thankfully all the giant flamewars and new policies didn't make anyone
> think twice about the users, as we already have 140 updates with a
> combined size of _over_ 750MB on x86_64, biggest 5 are:
yeah, over 750 MB where 584 MB belongs to wesnoth and openarena. So without
these two games it's only 166 MB and that's not that bad (also don't forget
> 6.2M wesnoth-1.8.1-1.fc13.x86_64.rpm
> 12M hanazono-fonts-20100222-2.fc13.noarch.rpm
> 48M xmoto-0.5.3-1.fc13.x86_64.rpm
> 260M wesnoth-data-1.8.1-1.fc13.noarch.rpm
> 318M openarena-0.8.5-1.fc13.noarch.rpm
> ...the last being particularly "nice", in that the package hasn't been
> updated for almost a year but now we get 2 300MB+ presents at once.
> Welcome to the new Fedora updates, much like the old Fedora updates.
why are you so scared about updates? Now we have some packages in Fedora that
are broken or outdated (for online multi-player games - you usually can't play
with other players when your version is not up2date, so it becomes unusable
even the update itself is enhancement only).
And what should maintainers do?
a) update only in F-14 so users will get broken F-13 and they will have to
wait another 6 months for new version?
b) wait at least two weeks after F-13 GA, so we won't have too much 0-day
updates? Does anyone really think lower 0-day update size is much cool than
delivering fixed packages?
c) wait until there is critical/security bug only? So bug reporter has to wait
three months for getting the fix for bug he reported? It'll just disappoint the
reporter and his conclusion would be that bug reporting is useless. Even when
the reporter can have the fix from updates-testing using the update command
pasted in the bug report by bodhi, there would be other users facing the bug
complaining about buggy fedora while waiting for the fix.
From my pov I'm happy we have maintainers working hard to get fixes and
features to our users. I'm fine with huge 0-day updates, because despite it's
not optimal, it's still much better than having broken packages. If there are
really a lot of 0-day updates it does not mean we have wrong update policy, it
just means release date should be postponed
> Hey, at least Kevin should be happy.
well, I am happy ;)
this is my first and last post to this flamebait thread
More information about the devel