ESR "fedora-submit"

[Eric S. Raymond]

Callum Lerwick <seg at haxxed.com>:
> For those of us who only joined the project in the last year or so,
> could you please point us to the original thread? WTF is this all about?

In a nutshell, I won't be a Fedora package maintainer if it means I have
to do special hand-work every time I ship a release.  The more general
problem is that the Fedora submission system doesn't scale (or at least,
appears to me not to scale) for maintainers with *lots* of projects.

In 2003 I offered to write tools to attack this problem by automating
the client end of release submissions.  I actually did write a script
that can enter bugs with attachments to a Bugzilla instance under
program control; it's been part of the mainline Bugzilla release
since, hmm, 2004 I think.

The next step was going to be to write a 'fedora-submit' wrapper
around that. Warren Togami liked the idea at the time, but seems to
have changed his mind since.

I trust my reply to Warren Togami will supply sufficient additional context.

> Whats the problem, and how exactly is it leading to the Imminent Death
> of Fedora Also Red Hat?

In 2003, when I first tried to attack this problem, Fedora was riding high.
Since then, a series of strategic blunders on the part of both Red Hat
and the Fedora project leadership have thrown away most of the advantages
in reality and perception that Fedora had at the time.  The best index of
their failure is the rising popularity of Ubuntu.

I view the failure of Fedora to address my scaling issue as a symptom
or indicator of a larger failure to do what it took to retain
community and developer support at the level it enjoyed in 2003.

When a developer of my stature and experience feels like he's being
forced away from a distribution he's supported for over ten years,
that's damn bad news for a distribution which, on the market- and
mind-share evidence, has already screwed lots of other pooches.

How do I put this delicately...oh, to hell with delicately.  If they
won't listen respectfully to a guy who helped found the entire
fricking open-source movement, my confidence that they'll pay the
attention they must to J. Random Developer is *nil*.  Warren Togami's
"on crack" remarks have just demonstrated that he has his head jammed
firmly up his ass, and (sadly) I have no evidence that the rest of the
project leadership is any less ingrown and cut off from reality.

I'm not predicting Imminent Death Of Fedora.  I'm saying that you
are running out of time and room to maneuver.  Your project leadership
has been displaying every sign of strategic incompetence since 
before 2003.  I honestly wish you luck pulling out of that dive.
-- 
		<a href="http://www.catb.org/~esr/">Eric S. Raymond</a>