question for board members

Matt Domsch matt at domsch.com
Fri May 7 14:47:28 UTC 2010


On Fri, May 07, 2010 at 01:28:08PM +0200, Michael Schwendt wrote:
> On Thu, 06 May 2010 15:07:51 -0700, Jesse wrote:
> 
> > On Thu, 2010-05-06 at 16:18 -0500, Mike McGrath wrote:
> >
> > > We've seen the project continue to grow but scale poorly.  Our packagers
> > > used to be able to do anything they wanted,
> 
> What does that refer to? The sheer size of the package repositories?
> 
> Fedora's Packaging Guidelines and Policies are so overwhelming, they
> have always frightened off potential contributors. And still, Fedora
> has opened the flood gates by focusing on "package review metrics" --
> quantity instead of quality -- which made some reviewers approve niche
> market packages, which in some cases are not even used by their packagers
> on a daily basis. Later, Fedora has made it easier for packagers to ignore
> bug reports by simply waiting for automated scripts to close them.

I don't see the "package review metrics" as focusing on quantity over
quality.  I see it as a way to reward, albeit merely with a "thanks
for stepping up", for people who review packages.  It's an otherwise
thankless task, and we have a backlog of package review requests, so
clearly someone is using those packages (if only we could have a legal
way to get a feel for how many).  Sure, it's possible that some
reviewers might push through with a poor review just to bump their
review stats, but I'd be surprised if that's a big problem.

> 
> > now have to follow a process.

The increase in packaging guidelines, and other bureaucracy, I think
reflects the "scale" problem Mike brings up.  I don't think we've
overwhelmed contributors with bureaucracy, but we have established
higher quality goals than in the past, both in packaging guidelines,
and the change in updates philosophy.  These may viewed as negative by
contributors who were used to the "anything goes, as often as you
want" philosophy, which has led to a poor end user experience in too
many cases.

When we had ~100 packagers, and ~3000 packages, if something broke,
one of the 100 would either fix it themselves or quickly find the
owner and get it fixed quickly. Those 100 kept up their quality due to
the BPB ("brown paper bag").  We're over 2400 packagers now, and
quickly approaching 10k packages (so quite an increase in scale), and
there's also less "shame" if a packager or package breaks something -
"the automated tools should have caught it".  Only they don't.

Personal responsibilty has gotten us quite a long way, but it isn't
sufficient at scale.  Guidelines, AutoQA, and "think before you push
an update to each branch" are there to help us get the quality
reputation back where it's historically been.

Thanks,
Matt


More information about the advisory-board mailing list