5

I've seen multiple meta issues where there's a discussion about a site policy or how to handle certain issues, and answers get marked as accepted. When this lines up with the wildly popular sentiment, it makes sense. But when the accepted answer is not the most popular one, it makes me wonder what "accepted" really means on meta.

Take, for example, this recently resurrected discussion about how to close game-rec questions.

It previously had an accepted answer with 14 upvotes. It now has a different accepted answer with 7 upvotes.

For the record, I'm on board with the new accepted answer. This is not an argument about that.

But it made me realize that whoever happens to be the one who asks the question on meta seems to have control over what gets marked as the accepted answer. Let's take it away from a specific case, and say someone comes in here looking for guidance on how to deal with some issue. They find a thread, and they see the following:

  1. An answer with a small handful of upvotes (say, 3-4) which has been marked as accepted.
  2. Another answer with a TON of upvotes (say, 20) which is not the accepted answer, and says something different.

At this point, they don't know which is correct. The accepted answer might be accepted because while not popular, some moderators laid down the law and the acceptance reflects that. Or it might be because the question asker happened to prefer that answer, even though it wasn't the popular and preferred one.

So what does answer acceptance mean on meta? I've seen this, which implies that moderators can accept answers to drive policy. However, that specifically talks about accepting the highest-voted option. What about when the highest-voted option is not the one that's accepted? It's not clear to me in those cases what the real policy really is. I have no way to tell if the question asker clicked accept, or if a moderator did it.

  • Is the assumption that an accepted answer on Meta always has moderator approval?
  • As a non-moderator, should I ever click "accept answer" in meta, or should I leave it for a mod to do?
  • Are all accepted answers on meta assumed to be the official policy?
  • If the accepted answer is not an indicator of policy, what is? How do I properly determine what the site policy is that was decided?
5

Accepted answers don't mean anything on meta.

Moderators can't change the accepted answer, so they are solely determined by the asking users and therefore can't be a reliable indicator of an answer being the official policy.

They have some meaning for support questions, where they can indicate which answer was most helpful in solving the problem. For feature requests I usually accept the answer from the SE dev that implemented it, if the request was successful.

The whole concept of accepted answers doesn't make much sense for large parts of meta, it probably should just be removed entirely, or at least the default sorting should ignore it.

  • Going to update my question, because if we can't go with accepted answer, I don't know what we go by. – Sterno Mar 19 '12 at 15:04
  • 6
    I shall now accept this answer, even though accepting it means nothing! – Sterno Mar 19 '12 at 16:45
  • 3
    I don't agree with this accept. – Resorath Mar 19 '12 at 19:46
3

Accepted can mean a lot of things; if it's a support or Feature Request post, it can mean "Okay, this post solves my problem" like it's supposed to, once the issue has been resolved. If it's a discussion, it really just means "this answers my question" (which it should!) or, much worse, "I agree with this".

To really judge what's important, look at the mod-only "status" tags to tell whether X feature did happen or if X bug was resolved, and for discussion, just look at the votes to see where the majority opinion lies.

The thing with Meta is that you have to actually read to understand what's going on; it's a discussion. Unless it's a bug/feature request with a solution posted, you can't just peek at one answer and know the full "answer"

  • Okay, so going to my specific example then.... which way should we be closing game-rec questions? Is it left for the reader to decide whether or not he feels @badp made a convincing argument to change, given that it is the accepted and most recent answer, but not the most highly voted? What if there's some other controversial subject and the vote count keeps flipping back and forth? Does that just mean "do the one you prefer because as a community, we can't make up our mind"? – Sterno Mar 19 '12 at 16:01
  • @Sterno well Badp's answer is accepted, logical and simply is how we've been closing those questions, I'd certainly say it's accepted in all significant ways. In this case it's a moderation issue as well, so the prevailing opinion of the moderators should be taken into account – Ben Brocka Mar 19 '12 at 16:10
  • Mana's answer is also logical, and he is also a moderator. And his has more upvotes. So why choose Badp's over his? What makes badp's the accepted answer? Because NickT is the one who asked the question, and therefore has control over what gets ticked as accepted? This is the entire thing I'm trying to understand. Your comment seems to boil down to "it's accepted because it's accepted". – Sterno Mar 19 '12 at 16:33
  • @BenBrocka except the other higher upvoted answer was also by a moderator, so I do see his point, which way are you supposed to go :P – l I Mar 19 '12 at 16:34
  • 1
    @yx. well badp's answer has 10 votes in 3 hours vs 16 votes in months :P It's also just....how things have been done, but that's harder to glean just by skimming answers – Ben Brocka Mar 19 '12 at 16:37
  • @Sterno The timeline is important for more than just votes also -- Mana's post seems to have basically flown under the radar. I happened upon it when NickT disagreed with other close-voters and I closing a game rec as OT, and after I had commented NickT bumped the post and Mark saw it and posted. The discussion was sort of stunted by the timeframe, and based on what I've seen we do close them as OT most of the time. – Matthew Read Mar 19 '12 at 17:04
  • @MatthewRead I think the main question I had was not really the one I asked... how do you determine site policy by reading meta? I figured accepted answers played a role, but they don't. The simple answer: You don't determine what site policy is by reading meta. You can sort of take the temperature, but pretty much the only way to be sure is to look at how things actually get done. – Sterno Mar 19 '12 at 17:10
  • @Sterno Pretty much. The big things are (or ought to be) in the FAQ (such as game recs being OT), and most of what's left isn't super important (close them as OT or NC, as long as they're closed!). – Matthew Read Mar 19 '12 at 17:32
  • @Sterno "How do you determine site policy by reading meta?" Well, if current policy is not clear, start a brand new question asking what our current policy is. Often Meta posts on policy are old and may no longer reflect the current opinion of the community, intent of SE or the direction of the moderation. Sometimes you need an update – Ben Brocka Mar 19 '12 at 18:40

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .