Tim Landscheidt <tim(a)tim-landscheidt.de> writes:
is there any metric defined to see how the bug solving
pro-
cess has improved over time?
No, but you gave us a really good one:
So is there any "scorecard" with more
detailed data than the weekly
report, i. e. minimum/average/maximum time to closure of
bugs/non-bugs, some nice visualizations, etc.?
I think that is a great idea. I've been thinking about doing something
like this, but you've really concrete-ized it.
Now to your other concerns:
The main reason I'm asking is today's two
mails regarding
bug #24000 ("Update rsvg so styling SVG images with CSS
works properly on Commons"). Reported more than a year ago,
the solution found early on was to update rsvg. I would have
suspected that to do that you execute an "rpm -Uvh" equiva-
lent on the server farm and close the bug.
As you suspect it isn't quite that easy.
Nevertheless this and a few other package upgrade bugs will are on the
list of items that Ops will probably be dealing with very soon.
I suggest that you use the newly created role of “Bugmeister” that I've
taken on. If you (or anyone else, for that matter) has a bug that
hasn't been dealt with, let me know and I'll do my best to get it to it
taken care of.
I'd like to have another method available than “bug the bugmeister” to
get these older bugs dealt with, but with almost 1600 open bugs, I'm
afraid I really don't have one (though, I'm open to any suggestions).
From an outside perspective, this looks like (much)
more energy is
spent on managing the bugs than actually squashing them.
The intent of the bug triages and other Bugmeister activities (“managing
the bugs”) is to get developers to resolve them.
Note that another way I'm trying to resolve the bugs is by committing
patches that people have submitted and getting them applied faster.
http://www.mediawiki.org/wiki/Special:Code/MediaWiki/91642 shows the
result of this.
I didn't review this closely enough, but it ended up getting reviewed
really well by other developers and the submitter was still paying
attention (partly, I think) because of the quick application of his
patch.
Let me know if there is anything else you think I should be doing.
Mark.