[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: packages missing from sarge

On Sun, May 08, 2005 at 12:36:16PM +0200, Adrian Bunk wrote:
> On Sun, May 08, 2005 at 08:45:21AM +0200, Andreas Tille wrote:
> > On Sat, 7 May 2005, Joey Hess wrote:

> > >bb
> > I did not checked your complete list but our most frequently used
> > programs at exhigition boothes.  It currently has no RC bug (the only
> > grave bug was solved two weeks ago.

> > So something is wrong either with your list of with the removal.

> That's a funny example of the result of the release team's focus only on 
> the RC bugs metric:

> This package had a more than six years old "normal" bug stating that bb 
> crashes on alpha (#32160).

> Last month, a second bug for exactly the same issue was sent with 
> severity "grave" (#304434). This second bug report included a trivial 
> one line fix for this bug.

> You might think the second bug made the situation better because it 
> included a patch for a more than six years old bug that made the package 
> unusable on 64bit machines. 

> But in the logic of your release team, the first non-RC bug didn't show 
> in their RC bugs metric while the second bug did.

Yes, it's called "garbage in, garbage out".  If people aren't going to file
bugs at the proper severity, and if package maintainers aren't going to
treat release-critical bugs with the appropriate urgency when they *are*
filed at the wrong severity, there's no way in hell anyone is going to know
there's a problem.

It's not the metric that's broken here.

Steve Langasek
postmodern programmer

Attachment: signature.asc
Description: Digital signature

Reply to: