Re: 2.4 Proposed Modules - 2 weeks left

[Alexander Larsson <alexl redhat com>]

On Tue, 6 May 2003, Peter Korn wrote:

> Jeff Waugh <jdub perkypants org> wrote:
> 
> > <quote who="Peter Korn">
> > 
> > > We already delay indefinitely for quality, right?
> > 
> > Absolutely not. We slash and burn. Whether than means slash and burn 
> > before the feature freeze, or a rare emergency slash and burn as we're 
> > closing in on a release, we do not delay indefinitely.
> 
> I was under the impression that the thing being "delayed indefinitely" in
> our discussion was the new feature/application, not the release overall. 
> Thus, if a candidate for being part of the desktop (which was never before
> part of the desktop) had major/massive quality problems, we would
> potentially delay indefinitely inclusion of said app/feature into the
> desktop - until such time as those problems were fixed.  Of course, if the
> quality problems were largely localized to one area of the app, you might
> "slash and burn" that portion.  But... that begs the question of who would
> do the slashing and burning, especially on any particular schedule in the
> face of no funding.  Hence we're back to the potentiality of delaying
> inclusion of an app indefinitely due to quality problems.
> 
> I didn't mean to suggest that a release of the *desktop* itself be "delayed
> indefinitely" in any of these cases (or for any other cases for that
> matter).

The *release* of the desktop is not "delayed indefinitely", but important 
development of the desktop is. Large development projects like a 
webbrowser have to get in at some point. It can't be finished, integrated 
and other parts of gnome adjusted to it while being an external unendorsed 
3rd party app.  

Of course, it has to have achieved some amount of quality before being 
considered for inclusion. This includes basic levels of code quality, 
maintainer willingness, integration, a11y, hig-compliance, etc and a 
willingness/plan to increase these.

Now, how does one decide when the quality is high enough in enough areas 
that the app/feature is ready for inclusion? One thing I thought we all 
agreed on is to not have hard feature blockers, which means we can't just 
block on a11y, while not considering other issues (feature importance, 
quality, etc). Of course, in the case of Gnome this "considering" 
basically consists of various people sending heated mails on mailing 
lists, and finally the release team gets to "interpret" this and come up 
with what they think is our decision as a group.

-- 
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
 Alexander Larsson                                            Red Hat, Inc 
                   alexl redhat com    alla lysator liu se 
He's a war-weary Amish filmmaker on the wrong side of the law. She's a 
strong-willed red-headed mermaid from a secret island of warrior women. They 
fight crime!