What's the reason for not accepting karma from Bodhi submitter

[Stephen Gallagher]

On Mon, 2012-08-20 at 12:28 +0000, Andre Robatino wrote:
> Julian Leyh <julian <at> vgai.de> writes:
> 
> > 
> > 2012/8/20 Richard W.M. Jones <rjones <at> redhat.com>:
> > >
> > > https://admin.fedoraproject.org/updates/qemu-1.2-0.3.20120806git3e430569.fc18
> > >
> > > I built the package, and I tested it.  Yet doing the right thing means
> > > my karma doesn't count ...
> > >
> > > IMO this change makes the karma system (even) worse than before.
> > >
> > > Rich.
> > 
> > If you submit it, it means you already tested and approved it. Karma
> > from you gives no new (quality) information about the package. It
> > should be tested by independent people.
> 
> That's not necessarily true. The packager might build for F16, F17, and F18,
> say, but not actually have all those systems to test on (even in a VM). So karma
> would in fact give additional information. IMO packagers should be treated the
> same as anyone else regarding karma. If they abuse the privilege, it can be
> taken away individually, the same as any other tester.


Furthermore, a not-inconsequential consideration is proxy-karma. For
example, I have in the past occasionally given karma +1 (and -1) on my
own packages from users that did not have Fedora accounts (or couldn't
manage to defeat the ogre that is FAS's CAPTCHA).

Not being able to add this karma on their behalf further reduces the
available pool of testers (or at least forces me to go find someone else
with a Fedora account and ask them to proxy it for me, thus resulting in
three layers of indirection).
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part
URL: <http://lists.fedoraproject.org/pipermail/devel/attachments/20120820/99bddf14/attachment.sig>