This happened to me: A question X was asked recently and was closed as a duplicate of a question Y. I faintly remembered these questions and when X came to me as a review audit question in the close vote queue, I voted to close it as a duplicate of Y. The audit told that I had done wrong; there was nothing wrong with the question. The question was indeed ok as such, but it was a duplicate. Moreover, I chose exactly the same closing reason that was used unanimously to close the question in the first place, and I failed the audit.

I know that audits can be very silly, but this sounds like an outright bug. Could this be fixed?

Here is the audit.

  • $\begingroup$ This is most likely due to the original post having high vote score (+6). It seems that the fix didn't. $\endgroup$ Oct 30, 2014 at 8:46
  • 1
    $\begingroup$ Failing a review audit once in a while is part of the human experience. It's like getting a hangover. If you get hangovers every day, something is wrong with you. But once in a while getting hungover means just that you had an awesome night. $\endgroup$
    – Asaf Karagila Mod
    Oct 30, 2014 at 9:11
  • $\begingroup$ This question is very close to this one. $\endgroup$ Oct 30, 2014 at 9:13
  • 3
    $\begingroup$ @AsafKaragila, I have occasionally failed audits before and I'm not upset by being stopped every once in a while to pay more attention (even if the audit is silly). But in this case the optimal behaviour (closing a duplicate as a duplicate) leads to failing the audit, so the audit is wrong, not just weird or disputable. If the audits are supposed to teach reviewers, this one is teaching the wrong thing. (Oh, and I've never had a hangover. I have fun in other ways.) $\endgroup$ Oct 30, 2014 at 9:54
  • 1
    $\begingroup$ This is really hard to fix properly. I have some ideas, but no promises. $\endgroup$
    – Shog9
    Dec 5, 2014 at 22:59
  • $\begingroup$ @Shog9 we can already search for questions that are closed or not with closed:[0-1]. So I assume there is a command for that. Now just don't take a closed question as an audit where the right outcome is that it shouldn't be closed. I wonder why it is so hard? $\endgroup$
    – wythagoras
    Oct 6, 2015 at 15:32
  • $\begingroup$ Not all duplicates are closed, @wythagoras. The correct solution here is probably to pass the audit if the reviewer elects to close the question as a duplicate, but that also requires actually recording the dup-vote which is... a little weird for an audit. Not impossible, but not trivial. $\endgroup$
    – Shog9
    Oct 6, 2015 at 16:34


You must log in to answer this question.

Browse other questions tagged .