On Monday 21 November 2005 09:39, Donnie Berkholz wrote:
> Ned Ludd wrote:
> | On Sun, 2005-11-20 at 14:45 -0800, Donnie Berkholz wrote:
> |>Our policy for X is that if upstream won't accept it, we won't either.
> |>Perhaps you'd be interested in adopting that and convincing the reported
> |>to get upstream interested?
> |
> | Your policy for X is somewhat questionable Donnie as it puts us in a
> | catch 22. You wont accept patches unless they came from upstream and
> | upstream wants some testing or to put it off till a later date..It's a
> | continuing heartache dealing with X when something could of been fixed
> | months ago.
>
> Upstream CVS is the location for testing, not distros. Distributions
> should have a _more_ stable version of packages than unreleased CVS, not
> less.
>
> In addition, we're in the business of packaging source, not maintaining
> source. Taking on maintainance of all the source we package is
> unrealistic and is not why I do Gentoo.

I think one should look at this as there being three kinds of patches:
- Those that add new features. If they are not upstream maintained they don't
  belong in the tree.
- Those that fix bugs. If the bugs are real and the patches are reasonable in
  quality and fix the bugs they help the users make things work.
- Those that do a mix of things. Only in extreme cases useful, but in general
  should be split out into the specific things they do.

Paul

-- 
Paul de Vrieze
Gentoo Developer
Mail: pauldv@gentoo.org
Homepage: http://www.devrieze.net