10k open Ubuntu bugs

Daniel Robitaille robitaille at gmail.com
Mon Apr 24 21:14:59 BST 2006


On 4/24/06, Sitsofe Wheeler <sitsofe at yahoo.com> wrote:
> On Mon, 2006-04-24 at 08:55 -0700, Matt Zimmerman wrote:
> > (ironically, this does belong on ubuntu-devel rather than sounder)
> >
> > On Mon, Apr 24, 2006 at 11:18:52AM +0100, Henrik Nilsen Omma wrote:
> > > Perhaps we should seek to make this explicit policy (such as via a TB
> > > decision). That way bug submitter will know that this will happen and
> > > volunteer bug triagers will be more confident in closing old bugs.
> >
> > This is absolutely appropriate, and most bug triage activity that I see
> > already works this way.  However, it would be logical to document it in the
> > wiki under HelpingWithBugs.
>
> This is interesting. I was chatting with an extremely on the ball Ubuntu
> dev in #launchpad the other day (and in fact someone on #ubuntu-bugs
> earlier today) about stuff like this. I do think that Ubuntu's bugzilla
> is a bit too "friendly" at the moment and is going to be crushed under
> the weight of "fix anything/everything" reports. Perhaps this discussion
> would be better off on one of the launchpad lists though (which still
> aren't on GMANE - http://www.gmane.org/ : )

In many ways I miss the days of the bugzilla when universe bugs were
quickly dealt by simply rejecting them.  That was really putting an
emphasis on the main vs universe/multiverse differences (and risks
when using these non-supported repos) for the users.  Now the line
between the repos is a bit more blurry, and the bug reporters expect
the same support and/or bug solving service for all the bug reported,
no matter where they come from since they  think they are all "ubuntu"
bugs.

And even during the Warty era when KDE bugs were as well ignored,
before Kubuntu was created that was easier still with only the main
ubuntu/gnome bugs that survived the initial bug triage.

I know all these bugs now belong to Malone, but maybe we should find a
way to create some sort of compartment system between the different
types of bugs we get.  I can't imagine how the developers feel when
they see these 10000 open bugs, and hundred of bugs assigned to them,
are all lumped together in one big pile, main and universe together,
ubuntu/kubuntu/xubuntu/etc together,  and they have to find the good
reports from the one to be ingnored and try to see patterns in the
problems submitted.    On the long run, I'm not sure the current
system is a sustainable one as we pick up more and more users, and
thus gather more and more reports.  First thing we know we will have
the same number of bugs reports in Malone than the Debian BTS since we
cover a very similar range of applications and packages, but with a
lot less of the manpower to deal with the reports.

--
Daniel Robitaille



More information about the ubuntu-devel mailing list