Example: https://stackoverflow.com/review/suggested-edits/2081208

Has 3 approve votes, and 2 rejects. It could have easily went the other way it seems.

Maybe the approval should come into effect not when 3 "for" votes are reached, but when the difference between pros and cons is 3? (or 2, or something). Just a thought.

  • 1
    This grates me as well. I've seen 100's of suggested edits which could have gone either way, and often go the wrong one in the end... – Matt May 9 '13 at 13:48
  • 8
    Then it could just end up consuming a ton of people's votes and in the end still be a toss up in which, eventually, one side gets 2-3 votes in a row. Assuming the votes are random with a 50% chance it will still happen eventually. The real problem is to address the issue of people just voting incorrectly. – Servy May 9 '13 at 13:57
  • 2
    That was a terrible edit, if it wasn't invalid then it was too minor. Grrr. – hayd May 9 '13 at 14:02
  • No, that will never end. Horrible edits can still be approved 5/3 then someone else will ask to make it difference of 3. Then horrible edit approved 6/3 and...... (endless loop here) – Shadow The Princess Wizard May 9 '13 at 14:15
  • 1
    Perhaps this should only apply to approving an edit? If 3 people reject, it gets rejected, regardless of the number of approve votes. However, in order to approve it you need 3 + number of reject votes to actually approve it. People mindlessly approving wouldn't be able to prevent it being rejected, whereas people reviewing correctly would be able to stall the approval in a majority of cases. I'm not sure if that's a better idea, or if it would be effective enough to be worthwhile, though. – Anthony Grist May 9 '13 at 14:16
  • As for the linked suggestion it's indeed invalid, rolled back. – Shadow The Princess Wizard May 9 '13 at 14:18
  • @AnthonyGrist It's better to fix the problem then to try to manage the symptoms. If you have people mindlessly approving stop them from mindlessly approving rather than trying to find ways of just making it harder to approve any edit. – Servy May 9 '13 at 14:29
  • @Servy I agree, but I'm also skeptical enough to think there's no way to actually do that; probably because I can't understand what motivates them to do so in the first place (I just don't think the badge is that big a deal). – Anthony Grist May 9 '13 at 14:34
  • 2
    @AnthonyGrist Well, there have been a lot of suggestions over a very long time, some that have been implemented, some that haven't. Adding review audits was by far the most effective of the changes made, and it has helped quite a bit. I'm quite confidant that there are a number of other changes that can be made to improve the system, all of which have been brought up before, including from scrapping the badges and the leaderboard entirely, creating an effective FAQ entry on how to properly review audits, and possibly some changes around what the requirements are to be able to review at all. – Servy May 9 '13 at 14:37
  • @ShaWizDowArd what's a little crazy is that this user made two edits to this question (and enjoyed 4 rep)... – hayd May 9 '13 at 15:47
  • @hayden yep but he also got one such edit rejected, then he suggested it again and got it through. – Shadow The Princess Wizard May 9 '13 at 17:53

I agree with @Servy that we shouldn't consume (an infinite? number of) people on this kind of edit voting, and I don't think "more eyes" would solve the problem. But perhaps, since it is easier to "pass" the review, if at least 2 (of 5) people reject then the edit should be rejected.

In my opinion this edit was bad and should not have passed through the review.... problems with the suggested edit queue have been raised before (a few times)...


Perhaps it might be a good idea to follow the suggested rule "2+ rejections is an automatic denial". Several thoughts:

  • It seems the preference should be to reject and not to accept — if the edit is that important, someone else will edit it anyway.
  • Further, if there are more errors (or bigger errors), my guess is that means it is likely that a rejected edit will lead to a better edit in the long run.
  • While this may lead to "But why isn't it a difference of 5?" this is an arbitrary system anyway. The aim is to make the edits more effective with the aim of improving the quality of questions as possible (which is why the voting weights have changed in the past). If it seems likely that this will make the site's review process more effective, then I think it a better idea to attempt, even if people complain.

You must log in to answer this question.

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