Proposed udpates policy change

[Matthew Garrett]

On Mon, Mar 08, 2010 at 11:27:04PM +0100, Michael Schwendt wrote:
> On Mon, 8 Mar 2010 21:59:29 +0000, Matthew wrote:
> > 1) Updates to stable that result in any reduction of functionality to 
> > the user are unacceptable.
> 
> Unless the fixes contained within an update are _more important_ than a
> dropped feature.
> 
> E.g. if upstream has removed some "functionality" deliberately, and
> upgrading to upstream's code is the only way to move forward.

In that kind of situation, I think the maintainer would need a very good 
reason to push this to a stable release. We've arguably been there with 
Thunderbird, and we saw how much trouble that caused.

> > It is the expectation of Fesco that the majority of updates should 
> > easily be able to garner the necessary karma in a minimal space of time. 
> 
> Your wording or FESCo's?

My wording, to be voted on by Fesco.

> In either case, I disapprove this strongly. I have failed to get bodhi 
> karma from bug reporters multiple times before. It is beyond my time 
> to pester bug reporters, so they would vote inside bodhi instead of 
> simply adding a comment in bugzilla. In many cases (ABRT generated 
> tickets), I cannot even get them to reply in bugzilla. I release 
> updates in return to
>  - problem reports found in non-Fedora places,
>  - crap I see in daily diffs I create for upstream projects,
>  - problems I find myself, which haven't reported by anyone else but
>    likely affect other users.
> I don't want such updates to be held up by artificial hurdles.

As I've said elsewhere, this is a problem that needs solving. But I 
don't believe that it's a problem that's best solved by allowing people 
to push directly to stable.

-- 
Matthew Garrett | mjg59 at srcf.ucam.org