Triaging aRTS bugs

[Matt Rogers]

On Mon, Jan 07, 2008 at 07:10:13AM -0800, Gary L. Greene Jr. wrote:
> On Monday 07 January 2008 04:43:56 am George Goldberg wrote:
> > On Jan 7, 2008 8:16 PM, Matt Rogers <mattr at kde.org> wrote:
> > > -----BEGIN PGP SIGNED MESSAGE-----
> > > Hash: SHA1
> > >
> > > On Jan 6, 2008, at 11:31 PM, George Goldberg wrote:
> > > > Hi all,
> > > >
> > > > I am trying to help with triaging out what cruft I can from
> > > > bugs.kde.org
> > > > and am wondering what the best approach to take is with 'arts' bugs.
> > > >
> > > > As far as I am aware (please correct me if any of this is wrong),
> > > > its unmaintained and pretty much dead. and, since phonon has replaced
> > > > it for KDE 4, its unlikely to be revived.
> > > >
> > > > Assuming the above is correct, I propose that each of the bugs and
> > > > feature
> > > > requests filed on bugs.kde.org for arts should either be reassigned to
> > > > phonon, or closed (depending on whether they are relevant to phonon or
> > > > not, since many of them will not be... since phonon != a sound
> > > > server).
> > > >
> > > > I am also volunteering to do this :-)
> > > >
> > > > Before I start, are there any objections or suggestions for
> > > > a better way of going about this?
> > > >
> > > > George
> > >
> > > The best approach to take for arts bugs is to do nothing. arts is not
> > > phonon and so moving those bugs from arts to phonon is not the
> > > correct thing to do IMHO.
> > >
> > > If you want to help triage bugs, the triage a maintained component.
> >
> > Yeah, that's what I'm mainly doing. But when I'm digging through
> > the older bugs in bugzilla there are always arts bugs cropping up.
> > I just thought that it might be possible, given its situation, to do
> > something about them once and for all.
> >
> > I agree that allocating them all to phonon is not the right
> > approach, but are we ready to just close them, or is
> > there still a big enough possibility that some features might be
> > implemented or some more bugs fixed to be worth leaving
> > them alone for now?
> >
> > > - --
> 
> Honestly, yes. There are going to be vendors working on the KDE 3.5.x releases 
> for some time yet, and they will likely be the ones to fix any arts bugs, so 
> for now leave it open.
> 

I'm not really seeing those vendors popping up to fix arts bugs though.
Here are some numbers:

Current number of open arts bugs: 223
Current number of open arts wishes: 29
Bug churn for arts over the last 30 days: 6 opened, 9 closed
Bug churn for arts over the last 180 days: 18 opened, 15 closed
Bug churn for arts over the last year: 27 opened, 22 closed
Bug churn for arts over the last 2 years: 85 opened, 89 closed

Number of arts bugs applicable to KDE 4: 1 (it's a knotify crash)


Now compare those numbers to the numbers for some other components:

Last 30 days:
Konqueror: 205 opened, 159 closed
KMail: 68 opened, 86 closed
Kopete: 34 opened, 40 closed
KWord: 3 opened, 6 closed

Last 180 days:
Konqueror: 828 opened, 451 closed
KMail: 380 opened, 340 closed
Kopete: 246 opened, 276 closed
KWord: 18 opened, 89 closed

Last year:
Konqueror: 1717 opened, 1158 closed
KMail: 816 opened, 858 closed
Kopete: 519 opened, 444 closed
KWord: 65 opened, 124

Last 2 years:
Konqueror: 3913 opened, 3199 closed
KMail: 1877 opened, 1598 closed
Kopete: 1363 opened, 1126 closed
KWord: 356 opened, 337 closed

I think we can safely say that there is absolutely no one working on
arts enough for us to have people care for triaging bugs against it.

If people want to help triage bugs, they don't need to go through just
the old ones. It's better to go through a product (or in some cases,
just a component) and provide reproduction information and test cases
that can be used to reproduce the bug. 

Thanks
--
Matt