Summary from yesterdays (mini) FESCo meeting

[Axel Thimm]

On Sat, Dec 30, 2006 at 12:50:26PM -0500, Jesse Keating wrote:
> On Saturday 30 December 2006 12:48, Axel Thimm wrote:
> > Just let me comment that when I explored becoming a contributor to FE
> > one of my most pleasant experiences when I checked the packages
> > submission procedure back then was the high quality of reviewing done
> > and the implied quality of the packages.
> >
> > Until this thread I wasn't aware of monolectical reviews and if this
> > would become a habit it would decrease the quality of packages let
> > through. Which I find a pity as one of the nicest parts of FE was the
> > quality of packages.
> 
> Whether or not the guidelines were regurgitated into the bug report
> has no bearing on if a valid and quality review was done.  None
> whatsoever.  All it does is say "this person copied/pasted some
> content from a wiki page, and possibly filled in some blanks".  It
> does not prove or disprove that the reviewer actually LOOKED at the
> package in question.  There is no way to tell that, without video
> proof of the review.  You have to trust your reviewers, and
> spotchecks go a long way toward that.  Just pasting content does NOT
> help the problem.

Well, in the same hyperbolic nature an "APPROVED" only says that a
person hit eight characters on his keyboard in the proper sequence.

We should stop assuming the worse from the reviewers and just guide
them to do their review properly. And even if you do assume bad
reviewer then the checklisting is the way to find what they missed and
why.

Let's try another approach: Other than the people having written the
review guidelines no one has memorized the list (probably the authors
didn't either) and will have to check the MUSTs somewhere, be it in
the wiki or his personal notes. If he is going to check the items why
not publicly in the bugzilla? It's zero effort in addition, unless the
reviewer didn't check the MUSTs at all, and that would be bad and
worth catching.
-- 
Axel.Thimm at ATrpms.net
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/fedora-maintainers/attachments/20061230/b33b792e/attachment.sig>