I recently created a Stack Overflow chat room to whip low-quality edit reviewers into shape.

It… worked. Most of them said they would be more careful about minor/invalid/vandalism edits and that they didn't know minor edits were bad: after my complaining, they improved.

That leads me to think that edit reviewers should be educated more.

Reject those you know are wrong

Is that ^ supposed to mean anything? Come on, there should be a Help Center article at the very least. A screen that needs to be manually clicked through that explains reject reasons and informs reviewers about how the system works would be even better.

  • They weren't called out before. Certainly not invited to a room about the behaviour. That's the difference. Otherwise even with all the warnings and messages they would continue because it was a system message. – random Aug 19 '14 at 0:33
  • @random Well, the warnings and messages could still be improved, even if that requires a "chat with an expert" type interfaces… :P – bjb568 Aug 19 '14 at 0:55
  • 1
    Would be nice to have some clarification over what is "too minor" (which has different unwritten rules depending on the age of the post) and if we should be rejecting tag-only edits that are technically correct but aren't dealing with the major tags on a post. – Troyen Aug 19 '14 at 1:00
  • 3
    Minor edits aren't that bad, and I see waaaay too many edits rejected on the ridiculous basis of "minorness". I'm sorry to see that you're going to make that worse. – Lightness Races in Orbit Aug 20 '14 at 15:17
  • 1
    @Troyen: Any technically valid edit should be accepted, IMO. "You didn't fix this other thing too" is a frankly stupid reason to reject improvement. – Lightness Races in Orbit Aug 20 '14 at 15:18
  • 1
    I agree (begrudgingly) with @LightnessRacesinOrbit minor edits are fine. If it takes ten edits to fix up a question, or just one, I don't care. As long as the edits are made, what's the problem? The edit queue is always under control. – thecoshman Aug 20 '14 at 15:23
  • 3
    Well I don't mind so much if we discourage users without edit rights from submitting edit suggestions that are really minor, as that does clog up the queue. However, rejecting such edits outright seems counterproductive, and going further by discouraging high-rep users from these edits (as I see on meta frequently) is just offensively stupid. – Lightness Races in Orbit Aug 20 '14 at 15:25
  • 3
    @LightnessRacesinOrbit 1) That's the rules, period. 2) Minor edits clog not just the edit queue, they also can be used to game the reputation system. 3) Bumping questions excessively is a problem. 4) There's no real reason why minor edits should get thru. Any positive effects are… minor, and reject-improving isn't hard. – bjb568 Aug 20 '14 at 16:20
  • 4
    @bjb568: 1) On a subsite dedicated to discussing and improving the rules, what sort of ridiculous logic is that?! 2) High rep users do not "game" the reputation system through edits; they neither want nor need to. 3) No, it's not. 4) I already addressed that: improvement is improvement, no matter how small. – Lightness Races in Orbit Aug 20 '14 at 16:27
  • 2
    @bjb568: 1) The rules do not define "too minor" clearly enough for you to use that as an argument against anything I've said; my interpretation of a valid "too minor" reject is one that rejects BrE->AmE. 2) So what? If the edit is valid, it's valid, and they deserve the rep. If they only made the edi to get rep, so what? That's why we have rep: incentive. If they're trying to game with bad edits, it doesn't matter: they're bad edits, so reject. Again, then, irrelevant argument. 3) Explain. 4) Yes, it still is. – Lightness Races in Orbit Aug 20 '14 at 16:35
  • 1
    @LightnessRacesinOrbit 1) "This edit is too minor; suggested edits should be substantive improvements addressing multiple issues in the post." 2) Nobody "deserves" reputation. 3) This bumped the post to first on my front page. 4) Ok, fine, if you really want to keep the edit, push the improve button. – bjb568 Aug 20 '14 at 16:40
  • 2
    @bjb568: 1) Right, weasel words. All vague and subjective. No specifics whatsoever, beyond "multiple" which yes I will choose to entirely ignore because that text is just wrong. "Minor", "substantive". 2) Everybody who adds value deserves reputation. That is the entire point....?! 3) You've still not explained why this is such a bad thing, only asserted so. If you hate the post so much, don't worry, because it'll get knocked off within minutes. 4) No, because I don't have time to improve it. You're still obsessed with this idea of all or nothing. – Lightness Races in Orbit Aug 20 '14 at 16:46
  • 2
    @LightnessRacesinOrbit 1) "Multiple" is the rule. You don't really have a choice, unless you want to be review banned. 2) Nobody "deserves" reputation. They happen to earn reputation thru their actions. (and thru roboreviewers) 3) Bad posts shouldn't be on the front page. 4) If you don't have time to review properly, please don't review. – bjb568 Aug 20 '14 at 17:47
  • 1
    @bjb588: 2) By definition, if you earn reputation through good actions, then good actions lead to deserving rep. 3) "Bad" is subjective, and somebody just improved a post, so bumping it may encourage others to finish the job. This is a good thing. Bad posts can never be improved if no-one ever sees them. 4) Stupid thing to say, totally ignores everything we've already discussed. – Lightness Races in Orbit Aug 20 '14 at 17:53
  • 2
    More than having reviewers reject valid edits (I'm absolutely with @LightnessRacesinOrbit there) I'd rather have the reviewers put more effort into checking tag wiki edits for plagiarism. More often than not high rep users aren't even aware of the fact that a tag wiki entry that is a 1-to-1 Wikipedia copy is not only bad but of questionable legality. – Christian Rau Aug 20 '14 at 21:38
up vote 25 down vote accepted

I agree. Whoever wrote those reasons should be doomed to answer questions like this under a pseudonym for all eternity.

But enough about the past.

Nobody is gonna read a help center article. If they were, they'd have already read it when they got notified of the privilege. So here's what I think the guidance should be:

  • Approve edits that clearly improve the post
  • Improve Edit when you can make additional improvements to the post
  • Reject and Edit to replace an ineffective edit with your own substantive changes
  • Reject edits that fail to improve the post... or that make it worse
  • Skip if you are not sure and want to go to the next suggested edit
  • 1
    Ok, when are the changes going to be made? – bjb568 Aug 20 '14 at 18:41
  • 20
    In 6-8 weeks, newb – Shog9 Aug 20 '14 at 18:41
  • 4
    Can the tag wiki/excerpt have an special line: Reject edits that fail to improve the post, make it worse or are plagiarization of other sources without proper attribution. – Braiam Aug 20 '14 at 21:53
  • Where do "too minor" edits fit into this? A minor edit may indeed improve the post, but they're not worth the time of edit reviewers so we want to discourage them by rejecting them. Maybe "clearly improve" should be "significantly improve". – Barmar Aug 21 '14 at 19:22
  • And then the Reject condition should add "or only correct minor errors that do not impair understanding (grammar, punctuation, spelling, capitalization)" – Barmar Aug 21 '14 at 19:25
  • 2
    They don't, @Barmar - I'm redefining Too Minor as "it's easier for me to just discard your changes and make my own edit than it is to approve it". Because no one could agree on what "too minor" actually meant. – Shog9 Aug 21 '14 at 19:27
  • I was about to post a "never mind", as I just noticed meta.stackexchange.com/questions/238333/… – Barmar Aug 21 '14 at 19:28
  • 2
    Uh… we're a little bit more than 6-8 weeks in, and the guidance text still sucks. – bjb568 Oct 23 '15 at 13:23
  • Yeah, had a lot of other changes in the pipeline when this was written - never got back to it, @bjb568. – Shog9 Oct 24 '15 at 2:43
  • 5
    Dalgas got this in today! – Shog9 Nov 6 '15 at 19:52

You must log in to answer this question.

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