[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-core-devel
Subject:    Re: call for volunteer on bug tracking
From:       Stephan Kulow <coolo () caldera ! de>
Date:       2000-05-01 17:33:04
[Download RAW message or body]

Matthias Elter wrote:
> 
> On Mon, 01 May 2000, Stephan Kulow wrote:
> (...)
> > Harri Porten wrote:
> > > Michael Matz wrote:
> > > > Hi,
> > > >
> > > > On Mon, 1 May 2000, Harri Porten wrote:
> > > > > > I need someone who tries to keep the bugs on
> > > > > > bugs.kde.org together. I don't have enough time
> > > > > > for this, but it needs to be done as more and more
> > > > >
> > > > > I'm trying to help but end up processing the easy, i.e. obvious cases
> > > > > only.
> > > >
> > > > I think he didn't meant fixing the bugs, but merely assure, that they
> > > > end up in the correct category.
> > >
> > > I understood it this way as well and wanted to stress the fact that even
> > > processing the reports can be non-trivial in some cases.
> > > Some reports are
> > >
> > > a) not related to KDE at all
> > > b) related to 3rd party code in KDE, e.g. kdesupport or kscd
> > > c) based on a misunderstanding on the reporter's side
> > > d) maybe valid but difficult to assign (e.g. sudden crashes. A memory
> > > leak _somewhere_ in KDE ? A buggy X driver ?)
> > > e) wrongly assigned (example: kdelibs maintained by Kalle ;)
> > > f) targeted at a developer who is not very active anymore
> > >
> > > a) and c) can easily be handled via "close". This leaves the reporter
> > > without an explanation, though.
> > >
> > > To help with f) we might consider sending a copy of all/some reports to
> > > kde*-devel ;) A volunteer might be willing to pick it up.
> >
> > Well, "close" is of course not the right tool for such things.
> > a-d should all be handled by friendly mails that we're no gods,
> > so can only fix bugs that are at least reproducable or otherwise
> > fixable. If you look at our bug system, it looks more like an
> > archive - which is _very_ bad for it's usefullness. Rather close
> > some fuzzy bug reports than keep the useful reports hidden!
> >
> > to e-f: the bugs aren't assigned to people but to packages. The fact
> > that the bug mails go to people is just to send them to someone who
> > might know better. I know, this didn't work out with such Packages as
> > kdelibs and kcontrol, but at least we tried. We can of course split
> > these packages further, but there is the question how someone reporting
> > should know if the bug is in kdecore or kdeui or KtabCtrl. For them
> > the bug is in "KDE" ;(
> 
> Bugs assigned to packages with more than one maintainer or without active
> maintainer should go to the mailinglists.
> 
> kmail -> kmail@kde.org
> konqueror -> kfm-devel@kde.org
> kdelibs -> kde-core-devel@kde.org
> kdemultimeda -> kde-devel@kde.org
> ...
> 
> By the way, is it possible to register more than one developer as maintainer
> for a package?
> 
Not directly with the bug system. But registristring mail aliases is
very easy -
something like developerswithkdelibscode@kde.org ;-)

Greetings, Stephan

-- 
... but you ain't had mine

[prev in list] [next in list] [prev in thread] [next in thread]