Re: Need for launchpad
Stephan Hermann, 2006-01-08 17:10:31 +0100 :
>> Yes, but nobody has proposed making those non-free services a core
>> part of Debian development. Yet Frans proposes we do so with
> This doesn't work, but using it as helper tool would help the
> project. Launchpad itself is not tight to one distribution.
Right, there's no package for a single distribution. Worse: there's
not even a non-packaged, install-by-hand tarball available. It's not
only non-free, it's also non-available. I'm not sure how it could be
of any help.
> So, the best way of improving is to use it, and if someone finds a
> bug or a glitch or an idea of making the workflow easier and faster,
> the best way is to file a bug or kick the developers :)
Let's assume the "non-available" problem eventually gets fixed.
Debian gets and runs a copy of Launchpad. Now the best way if someone
finds a bug is to fix it himself, and then send the patch. That way,
the upstream developers can just review and apply, and the user can
apply locally before even that. That's my idea of an easy, fast *and
efficient* workflow. Unfortunately, if we're still in non-free stuff,
we can't use that workflow, we're stuck waiting for someone (whose
boss may or may not have the same motivations as you) actually fixes
the bug and releases a new version, which you'll have to upgrade to
when you find the required time.
Good workflows don't have big blockers, they allow the flow to go
around small blockers locally.
Seems to me, the only sensible thing is for people to know if they kill a whale,
they've got a dead whale. -- Adam, in Good Omens (T. Pratchett and N. Gaiman)