One of the users at History SE, proposed a system where moderators are granted only 3 close votes, as opposed to their current power to unilaterally close questions. This would have the effect of requiring two moderators, absent other user close votes, to close a question. I'm assuming that the reason moderators have this power is for efficiency reasons, but has this type of alternative system been implemented before?
What is meta?
- Any Stack Exchange user can ask a question
- The community provides support, votes on ideas, and reports bugs
- Your voice helps shape the way Stack Exchange operates
To the best of my knowledge, this system has never been considered on any Stack Exchange site. I'm sure there are plenty of forums, BBSs, or Q&A sites that have tried it.
Moderators on Stack Exchange sites have nearly unlimited powers. There are limits - they can't migrate questions that are more than sixty days old, for instance, a relatively recent change. They can't see lots of things, like detailed voting data, that the devs can see. But in general, they've got the power. And that's a good thing.
To use a well-worn phrase, moderators are Human Exception Handlers. Cleaning up messes is what they're here for. Not real sure a question should be closed? Leave it up to the community, wait for five 3k users to chime in. A rant about Google's robots.txt-based world takeover? Kill it with fire.
Moderator powers can be abused, in the small or large. If that happens, often enough, it'd be time to consider measures, like the one you're discussing. I've never seen it, though. Moderators, maybe by dint of going through an election with dozens of contenders, do amazing jobs, and don't screw up, that I've seen. If they did - if they're regularly closing questions that don't deserve it, or otherwise abusing their powers - then the thing to do it recall them. Revoke power from the berserker, leave it in the hands of those who wield it well.
I found that getting the power to unilaterally close changed the way I think about closing questions. There is no longer any wiggle-room to evade the responsibility, I can't just say "there were four other users who voted to close, I'm not responsible that we all made a bad call". When I close as a moderator I am fully responsible for my decision, and I should be sure that my action is the correct one.
This means that I avoid unilaterally closing borderline questions where I'm not sure if they should be closed. Unless they cause too much noise or trouble, I leave those to the community.
But I see no harm in moderators using their close powers if they are sure about it, they are still members of the community and should not hesitate to act if it is appropriate.
Of course moderators will get it wrong occasionally, most of them are only human and might misread a question and make a mistake at some point. But the community makes mistakes too when closing, requiring five votes does not ensure 100% accuracy. If moderator closings are reversed rather often, then there is obviously a problem. Either the community is split on the interpretation of the site rules, or there are no sufficient rules for this case. Those are all cases that need some kind of meta discussion anyway, just making moderators less powerful doesn't solve these problem.
The power to close is also by far one of the least dangerous powers, it can be undone by the community if they think the moderator was wrong and it is very visible. Moderators have access to far more dangerous tool, if we trust them with those why shouldn't we trust that they use their close power responsibly?
Moderators should not usually vote to close a question, even in situations where they would have done so before they became moderators and gained powerful mod close votes. Thus, the solution to moderators being so powerful is for them not to use their power without due consideration and deference to their communities.
Usually, at least as far as I have seen, moderators of Stack Exchange sites act within their limited role and try to defer to their communities. That is, the solution is currently implemented pretty well.
If a situation arises where or or more moderators does not do this, then this is not a technical problem that would call for a technical solution; instead, the solution is for them to use their power more appropriately.
If a moderator acts inappropriately, the solution is for us to politely but firmly object, explaining why we think their behavior was a problem. If the matter becomes serious (or starts out that way), the solution is for members of the community to contact the community managers about the matter.
In my opinion, moderators should only vote to close a question when one of the following situations applies, or perhaps something clearly similar. These criteria are subjective, this is my opinion (though it is distilled from things I've heard moderators say), and it's certainly possible for people to disagree about whether or not a particular closure fits into any of these situations. On the other hand, many possible closures clearly do fit, while many others clearly do not.
This is a list of situations where it is not a problem for moderators to cast a close vote:
Outside those situations, moderators should be extremely reluctant to cast a close vote. That's the solution to the "problem" of moderator close votes being effectively five times as powerful as a normal close vote. They should be used far less than one fifth as often.