Five build fixes a day

[Martin Pitt]

Colin Watson [2011-09-13  9:55 +0100]:
> I'm seeing good progress with FTBFS fixes being applied by Debian, so I
> expect many of these to turn into syncs.  Sure, some will turn into
> merges, but it's *not* simply pushing the whole pile of work off for
> later.

I checked some five random FTBFS fixes that were applied in oneiric
recently. One was a native package, two were forwarded to Debian, two
weren't. So there is indeed some good progress here, and hopefully
significantly fewer merges to do than I feared.

> > I'm actually in that camp. I am not convinced at all that fixing all
> > FTBFS bugs in universe is time well spent. We don't even have enough
> > manpower to fix even the worst bugs in main. We have only 1/10th or so
> > of the required manpower to keep the whole universe in good shape, and
> > every manday spent on random universe packages makes our actual
> > products worse.
> 
> Please find me a real Ubuntu user who doesn't install some packages from
> universe.

We all do. But ¬(fix all universe) ≠ (don't fix anything in universe)

I. e. I think there is a lot of cruft in universe which nobody ever
looks at or uses. E. g. I just removed some hildon packages which
haven't had any upstream or Debian/Ubuntu attention for several years,
it's just a dead project. We also removed a lot of GNOME panel 2 based
applets which don't work in GNOME 3 any more; porting them would be an
inordinate amount of effort, and for many of them we have better
solutions available now (e. g. mail notification applets and the
like). IMO we are much better off when we remove broken packages that
aren't maintained upstream or in Debian any more than spending lots of
time trying to get them to build again, just for the sake of it.

> I think we are too insular a group to be able to evaluate removals
> very well in general.

I agree wrt. the purpose of the package. I think we can evaluate
pretty well if a package is still alive or dead upstream and whether
it's just hopelessly obsolete (linking against ancient libraries,
etc.)

> Honestly, though: most of the build failures are trivial

Yeah, maybe I have just been unlucky. I looked at
vala-dbus-binding-tool (20 mins), valatoys (gave up after 25 mins),
libsmbios (15 mins), honeyd (had a quick look, but far from "easy").

> In the case of the 'ld --as-needed' bugs in particular, a lot of
> those have never been filed in Debian, and since the packages in
> question are primarily maintained in Debian, nobody with maintenance
> responsibility ever heard about the problem.

Right, for those in particular the fixes should be sent to upstream
right away, not to Debian, as it's not a problem that actually affects
Debian (yet).

> I also maintain that the amount of build failure noise is a problem in
> itself, and I really think it will take a lot less effort to keep that
> list at or near zero than it does to get it there.

OK, let's agree to disagree here.

Martin

-- 
Martin Pitt                        | http://www.piware.de
Ubuntu Developer (www.ubuntu.com)  | Debian Developer  (www.debian.org)