Re: Is Ubuntu a debian derivative or is it a fork?
On Thu, Jun 02, 2005 at 09:17:46AM +0200, Andreas Tille wrote:
> In your first mail you wrote about "mass-mail Debian maintainers" in the
> second mail you turned my request to file wishlist bug reports against
> single packages into "mass-filing bugs in the BTS".
If all of the patches were to be filed in the BTS, automation would be the
only feasible way to do it. It has been said that it is too much of a
burden for Debian maintainers to process the patches, and Ubuntu currently
has a miniscule number of developers compared to Debian.
> I guess it is a problem of my bad English but I mean if there is a problem
> in a package file a bug report against this package.
Not every change is an indication of a problem. Consider, for example, the
hundreds of Python transition patches. It makes more sense to make the
changes, and make the patches available to Debian, than to ask the Debian
maintainer to do the work for us by filing a bug report.
The same logic applies to many bugs as well. Would it really be better to
have an open bug report in debbugs, than a patch on people.ubuntu.com?
> BTW, there is a debian-python list
> where transitions oy Python packages can be discussed, if something is
> broken in the current Debian packages.
Debian uses python2.3 by default. This isn't broken; it's just the way
things must be in order to release Sarge. Meanwhile, Ubuntu has made a
release with 2.3, and is preparing for the next release with python2.4 as
default. When Debian makes the same transition, hopefully the patches from
Ubuntu will save some time and effort on the part of Debian maintainers.
> I see no reason to even mass-file bug reports if there is something wrong
> in a set of Debian packages. I would be happy if people would stop
> assuming that DDs who are not involved in Ubuntu personally per definition
> are criticising Ubuntu.
I don't think that I've made such an assumption, though there are several
unequivocally critical voices in this thread.
> This is at least not the case for me personally and I just try to enhance
> the situation for both sides. If there is a reason to mass-file bug
> reports than I see no excuse for not doing it. It might be that I'm
> simplifying things but often simple solutions are these who are hidden
> very hard. I just think that filing a megabyte patch to some Debian
> maintainer via a script is not really the solution. There is something
> wrong *before* this patch was created. This *something* might be caused
> technical or social (or both) reasons.
I know of no reason to mass-file bug reports, except that some people insist
that the best place to publish Ubuntu's patches is in debbugs. I disagree
with that position, myself.
We are doing what we can, with the resources available to us, to make our
work available to Debian, through the patch publishing mechanism, and
cooperation with Debian teams. If there is a different approach which could
be implemented using the same resources, I am willing to listen to such
proposals, but we have very limited developer resources compared to Debian.