[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Upcoming NMU; svn commit access for it is open for you



On Sun, Feb 01, 2004 at 03:47:42PM +0100, Josip Rodin wrote:
> On Sun, Feb 01, 2004 at 04:43:56AM +0100, Jeroen van Wolffelaar wrote:
> > > > I just created a subversion repository with the latest lintian release
> > > > in it.
> > > 
> > > We need to convert the CVS repository to it, merge in the past releases
> > > (I have tarballs from all but two or three of them, Shaleh forwarded them to
> > > me and I've kept all that I released) to make it a full-blown repo.
> > > 
> > > Note that space on svn.d.o was already allocated for Lintian...
> > 
> > If you can make those CVS things available, I could help with the
> > conversion (converted 2 weeks ago our 3300+ revision cvs repository on
> > some other project).
> 
> Great. I'll copy the tarballs on a removable drive and take it to a
> Place With A Fat Pipe and send the URL, probably tomorrow midday.

Okay. I will hold of applying further patches, just in case it turns out
it won't be possible to reapply current subversion history.

Maybe it's a good idea for others (Frank) too, just to prevent extra
work for me if I need to fix those manually per commit...
 
> > I've noted that you already wanted to start a more communtiy-managed
> > lintian (you started this very mailinglist, you had the svn repository
> > created, etc). I think this is a good start, past history would be great.
> > This moved to alioth for garanteed continuity is also a better solution,
> > I'm sure on alioth some subverion repository can be imported.
> 
> Yes, I have no interest in usurping all control over Lintian, that was most
> certainly never the intention :) I ,,rescued'' it from one inactive
> maintainer (hi Shaleh!), it's way past time for others to more actively chip
> in. You will note that Colin Watson, Frank Lichtenfeld and many others have
> already submitted gobs of patches, and I hate it when my business IRL is the
> reason why those patches aren't reaching the archive in time.

A way of collaborative maintainance that works well for some other
packages, is apache's way: Have the 'Lintian maintainers
<lint-maint@d.o>' as maintainer (or alternatively 'Josip Rodin and
others <lint-maint@d.o>'); and you and probably Frank Lichtenheld and
maybe Colin Watson too as Uploaders (= Co-maintainers officially, but
meaning 'the current lint-maint group in this case).

Bugmail etc will then be directed to this list.

What is needed is then only a policy on who and when and with the
consent of who a release can be done. I suggest something like:
- after 2 weeks of the latest not-uploaded patch is in subversion, there
  is branced for a release
- testing starts on it, at least two 'changes reviewed & testing OK'
  reports from maintainers needed, and for example 2 days. Only real
  bugfixes are accepted, and no veto's for any reason
- Upload, who in fact the upload performs is not very relevant/important.

And some kind of conflict/disagreement resolution. I propose the
benevolent dictator model (with Josip Rodin as the dictator), because it
is the easiest, and if it doesn't work out, it can always be changed.

> Note also that this list is not limited just to Lintian -- with CVS/SVN
> commit logs and/or diffs being sent to it, we can improve both tools at the
> same time.

You are referring to Linda I assume? It'd be best if both could be in
the same repository then, as you can then commit one fix/improvemt to
linda and lintian in one commit. But... then the linda cvs (if existent)
is needed too for the import. I could simply import all old releases if
that is not available. But, what does the linda maintainer think about
this?

Steve Kowalik, are you reading?
 
> > Maybe it's a possiblity if you, Josip, upload from the repository, so
> > bugs are closed rather than tagged fixed?
> 
> Yes, that would be just fine. Proofreading already committed patches via
> ViewCVS requires less man-hours than inserting them myself.
> 
> Remember also to add regression testing for additions.

That is a very good suggestion.

--Jeroen

-- 
Jeroen van Wolffelaar
Jeroen@wolffelaar.nl (also for Jabber & MSN; ICQ: 33944357)
http://Jeroen.A-Eskwadraat.nl



Reply to: