I've spent the last week or two observing and trying to help out with the
backlog on [[Category:Articles to be merged]]. Unfortunately, it looks like
the backlog is rapidly spinning out of control -- I haven't actually kept
track, but I have a sneaking suspicion that there are many more merge
requests added than are serviced every day. Right now the backlog is a
little under 10,000 articles, and I would like to see if we can't brainstorm
a new way to deal with backlogs, because our current methods aren't dealing
with our worst backlogs. (Our worst backlogs, ironically but expectedly, are
the ones users are least likely to start tackling.)
I was just thinking about the math, however: if 100 Wikipedians volunteered
to service one merge request per day, or 50 Wikipedians volunteered to
service an average of two per day, etc., we might make significant headway
into the backlog in three month's time. Of course, there would be new
articles to be merged, but ...
... anyway, it might not even involve actually performing the merge. A lot
of old merge requests are merges that have not gained consensus and should
just be removed. Also, since a merge request typically entails two merge
templates, it's likely we have "only" somewhere on the order of 6000~7000
actual requests. So if 100 Wikipedians, every day for a few months:
1) evaluate one merge request
2) check if the merge has consensus (or use their own discretions)
3) perform the merge
Maybe we'll make some progress clearing the backlog. (In fact, item #3 might
be as simple as copy and pasting and slapping {{cleanup}} on it to have it
checked over for redundancy and style, but of course, we'd have an even
bigger cleanup backlog then. :-) ) Any thoughts?
--
theProject