New Package Process

[Thorsten Leemhuis]

Am Mittwoch, den 27.04.2005, 17:30 +0200 schrieb Michael Schwendt:
> On Wed, 27 Apr 2005 11:00:23 -0400 (EDT), Greg DeKoenigsberg wrote:
> 
> > "Failing to build on one arch = blocker": If we're going to commit to this
> > policy, we need to ensure that every developer has access to every arch.  
> 
> And even then, you expect volunteer packagers to spend time on debugging
> problems with architectures they don't have interest in.

And even is they had, I still suspect that all volunteer packagers have
enough knowledge to fix packages on all those archs (I, for example,
don't have that that much programming experience to do this). 

> [...]
> I would really prefer if "architecture specific Fedora community
> developers" filled the role of package co-maintainers.

++ 

And even more: If one package does not get fixed in lets say around (4|
7|10|14|"your vote here") days, publish it for the other archs if it
does no other real harm for the project. 

>  Else we would play
> the "if it builds, publish it" game and offer something, which has not
> been tested at all.

Well, testing the important packages might be possible. But I suspect
"architecture specific Fedora community developers" can test all
releases and updates of Gregs "... we scale this project to a million
packages?  Because that's the goal". 

Maybe we should use the testing-repo more. Verified packages could get
moved to the normal repo after 4 (or maybe 7) days, unverified packages
will get moved automatically after 28 days (for example).

Just my 2 cent

-- 
Thorsten Leemhuis <fedora at leemhuis.info>