Tim Starling wrote:
So, what should be next ancient body dredged up from Bugzilla and resurrected by Platform Engineering?
In many ways, I view these two bugs as similar to the departed bug 189 (in terms of being old, annoying, and seemingly having had some progress made toward them, but never having gotten finished):
* https://bugzilla.wikimedia.org/show_bug.cgi?id=9890 "Reasonably efficient interwiki transclusion"
* https://bugzilla.wikimedia.org/show_bug.cgi?id=3311 "Automatic category redirects"
While bug 3311 is apparently slated for GSoC 2013, surely we've learned our lesson here year after year.
And bug 9890 could save thousands of man-hours per year. Given the concern that wiki editorship is on the decline, maximizing editor time by minimizing duplicate effort seems like a pretty clear win.
In my mind, I think the winner would be a proper wiki configuration graphical user interface, though. This would make MediaWiki far more attractive and robust, it would help with project sovereignty (by allowing Wikimedia stewards to implement requests for individual Wikimedia wikis and by allowing local wiki bureaucrats to administer their own wiki without the possible need for a shell user/FTP access/whatever), and it's the kind of undertaking that many outsiders are uninterested in pursuing themselves, but that would benefit both Wikimedia and MediaWiki greatly.
* https://bugzilla.wikimedia.org/show_bug.cgi?id=26992 "Implement configuration database aka configuration management aka no shell excuse (tracking)" https://mediawiki.org/wiki/Requests_for_comment/Configuration_database
MZMcBride