hard time finding packages really for review

[Patrice Dumas]

Hello,

I have seen floating around the idea that there was a need for reviewers 
for old reviews. So I started with the packages in FE-NEW with no activity
in eight weeks:
http://fedoraproject.org/wiki/Extras/PackageStatus?highlight=%28package%29#head-6ebf09953da14bdd17d75280dba5b4cffb941077-3
but I had a hard time to find a package really waiting for a review. In
the end I ended up looking at all of them, and here are the statistics
(I removed kmobiletools, accepted closed):

7  packages are stuck in review for various reasons
11 packages are stuck because they are waiting for submitters
10 packages were stuck with a need of a reviewer

So we have only 10 packages for 28 (about 1/3) that are really waiting 
for a review. This makes searching for a really open review quite 
painfull. So I think we should do something to be able to discriminate
the packages 

* stuck for various reasons except needsponsor, including 
because of packaging guidelines blocking the review like recently some
php reviews.
* the packages stuck because the submitter has to take an action 
but hasn't moved for, say, one month.
* leave the other packages under review as is.

a possible proposal could be along blocking special bugs that would 
change the background color in the trees like 
https://bugzilla.redhat.com/bugzilla/showdependencytree.cgi?id=FE-NEW&hide_resolved=1
to a given color. These colors could take precedence over the status 
(like NEW, ASSIGNED...). It would be nice if it could also be seen on 
top of the bugreport, but if it is only in 
Bug xxxxx blocks FE-SUBMITTER-ASLEEP FE-NEW
it would be right too.

I don't think it should be a must, but something nice to have. some
members of the fedora QA team, or anybody else, could for example look
from time to time on old reviews and make the blocks.

I don't care about that proposal, it is just an idea thrown, it might
not be feasible, but I think something should be done to help find
where reviewer can really act. 

Lastly, there could be a procedure for AWOL/MIA submitters similar than
the one for AWOL/MIA maintainers, to allow submitters to replace 
AWOL/MIA submitters and restart a review. This is allready done on a
case by case basis, but I think a policy wouldn't hurt here, and having
a classification of review stuck by submitter could help to find the
AWOL/MIA submitters.





Here are the details if you don't trust my numbers :)

Review stuck 6:
disagreament:
fnord freedt

licence issue:
acx-kmod acx-kmod-common

wondering about whether it is a good idea to include in extras or not:
alsa-oss phpBB

needsponsor 1:
gq

waiting for submitter 11:
smixer smokeping sparse gitweb mondo lurker libpri magic socat
fxload kdiff3

Waiting for review 10:
xmms-musepack transconnect ardour pyscript Wcl linux-wlan-ng gpc
vdr-osdteletext vdr-subtitles sqlgrey

--
Pat