Re: Filing Hurd-porting patches in the BTS or upstream? (was: where do...)
On Sun, 19 May 2002, Marcus Brinkmann wrote:
> On Sun, May 19, 2002 at 01:51:07PM +0200, Eduard Bloch wrote:
> > > > I won't make such changes behind the upstream's back.
> > >
> > > What do you mean with that?
> > Exactly this. Bugfixes are okay, new features too as long as they do not
> > break things. New operating systems - no go, since JS generaly cares
> > about portability issues.
> Maybe it is my fault that it happened this way.
> I think I encouraged everyone to file every bug first as Debian bug report,
> and leave it to the maintainer to decide what happens from there. This is
> because it is simpler for us (we know the Debian bug address for each
> package), and also because I didn't want to "circumvent" Debian on this
> matter, trying to make it possible for everyone in Debian to take part into
> the development of Debian GNU/Hurd, at least for the packages they
If something is in Debian, the bug should be filed in the Debian BTS, period,
end of story.
It's the maintainer's *job* to then sort those reports, and filter/forward
them upstream. It's just that some maintainers don't know this, and would
rather close the bugs in Debian's bts, and insist that the filer do the
maintainer's job, and file it upstream.
> Independent of this specific bug report (I really only skimmed over it),
> I am starting to reconsider this, and file porting problems upstream first.
> Although the other GNU/Linux ports file them as Debian bugs AFAICS, it seems
> that the Hurd-specific nature of the patches, and often the volume of changes
> needed, is outweighing in costs the benefits listed above. But the Debian
> BTS is really nice, and I like to use it to keep track of my porting work.
No, don't do this. Keep filing them in the BTS.
I really disapprove of misuse of the bts. Maybe we need a Best Practices
To UNSUBSCRIBE, email to firstname.lastname@example.org
with a subject of "unsubscribe". Trouble? Contact email@example.com