http://fedoraproject.org/wiki/TomCallaway/SecondaryArchitectures

[David Woodhouse]

On Wed, 2007-07-11 at 19:45 -0400, Christopher Aillon wrote:
> The term package is overloaded here.  In general the failures are in 
> upstream packages; not the Fedora package glue.
> 
> While we want to and should encourage packagers to fix bugs in upstream 
> code, dealing with bugs that are not part of an official Fedora release 
> should never be a requirement for Fedora packagers, IMHO. 

(That's actually a topic on which I violently disagree. If the person
responsible can't deal with bugs and properly maintain the package with
any degree of quality, we're better of not having the package at all. I
understand that some people value quantity over quality though.¹)

But still, it wouldn't be a _requirement_. The only requirement would be
that they _look_ at the failure and decide whether they care or not.
Hell, I suppose even that isn't required -- we wouldn't _even_ force
them to look, if they _really_ don't want to even look at why their
package failed to build. They could just follow the link to bugzilla and
click 'submit' without putting _any_ more information into the template
they're given, then ship the package anyway.

It just isn't that hard. (Regardless of my opinion of whether we
actually want that kind of package{,r} quality in Fedora.)

-- 
dwmw2

¹ If the sponsor were expected to co-maintain, that might help.