Brion L. VIBBER wrote:
It does seem a potentially controversial point... I think Magnus stuck it in specifically so he could grant himself developer priviledges and fix a problem with the database that was causing trouble (see below). I'm not sure whether this is intended as a permanent feature or if it's a entirely good idea.
Wikipedia is based on trust. If we can't trust sysops, we're in big trouble anyway. :-)
I was opposed to the general ability to let people (even sysops) execute arbitrary SQL commands via the web, not because of trust per se, but because of the high risk of an accident. It makes me personally very nervous when I have to type in raw SQL, and so I imagine it's not a good thing for anyone to be doing, except for developers and then only for specific and emergency-ish reasons.
Sysop status should be available to just about anyone who asks, and the privileges of the sysop should be so minimal as to justify this policy.
The main thing we want to prevent is the notion of an "elite" who directs the wikipedia and gets to say who participates in driving the content *through the use of special powers*. The rare exceptions should be agonized over, and should be restricted to people who just refuse to even try to get along with us.
I think that all sysops and especially developers should work really really hard to make sure to NEVER use any special powers in a content argument. It's just not fair, and not in the spirit of wikipedia consensus.
For example, I'd be pretty pissed at a developer who made a raw SQL command to delete another user, or to delete an article that was in contention. Yuck.
--Jimbo