4

If i find a pending edit to a post while browsing, but have already reviewed 20 edits that day from the queue, i'm unable to review that edit.

I'm pretty sure that the reason for limiting to 20 reviews is so that people don't get burned out of reviewing. That's a good idea.

I think though, that if you come across an edit while browsing, approving or rejecting it shouldn't burn you out.

Doing this wouldn't even have to count as a review, so noöne could review 21 edits in a day, and it would just be altruistic. Or else, it could count as a review.

Thoughts?

5

1 Answer 1

1

Well, this is old, but since it's unanswered...

The reason for limiting the amount of reviews wasn't burn-out. It was an early line of defense against robo-reviewers - limiting the amount of harm they could do in a single day.
Witness also the fact that if the queue exceeds a certain size, the amount of possible reviews per day goes up from 20 to 40. If burnout were the reason, we wouldn't get double the review slots.

The request makes sense though. I sometimes encounter posts that I want to edit, but that have an edit pending. And when I'm out of reviews, I can't do anything - I can only wait until the pending edit gets Approved or Rejected.
So there would be an advantage to allowing edit reviews from within a post, in excess of one's daily allotment.

However, it should not count towards the Reviewer and Steward badges, otherwise it provides robo-reviewers with a way around the 20/40 review restriction.

10
  • "otherwise it provides robo-reviewers with a way around the 20/40 review restriction." -- You really think a robo-reviewer would seek out posts with suggested edits for the badge? That's like, so much more work that i can't even imagine it.
    – Scimonster
    Aug 31, 2016 at 13:59
  • @Scimonster Hm yes, robo-reviewers might well be too lazy to do that... until somebody builds a script to find pending edits. Such a script would be easy to make, since edit suggestions have increasing ID's in their URL. Aug 31, 2016 at 14:04
  • Let's burn that bridge when we come to it. I think doing something like that would be ground for being blocked (or at least review-banned).
    – Scimonster
    Aug 31, 2016 at 14:05
  • (1) This would be nicely analogous to the 30/40 rule for votes. Say you can do 20 (or 40) suggested edit reviews from the review queue, plus 10 more that you find on your own. (2) Why are we concerned about people being able to review more than the normal 20 (or 40) in a day? Just because suggested edits found "in the wild" are guaranteed not to be audits? It seems like, if there's a feasible way of finding pending edits, then the robo-reviewers would do that and breeze through 20 (or 40) suggested edits that are guaranteed not to be audits. I.e., they can do that now. Sep 7, 2016 at 4:11
  • @Scott I don't think robo-reviewers are concerned about audits.. I've seen quite a few who approved everything that wasn't an audit. The cap is badly needed. I know of at least one reviewer who openly admitted to not care about site quality at all, only about the badges. Sep 7, 2016 at 7:42
  • @S.L.Barth: Well, doesn't that prove that they're sufficiently concerned about audits that they invest the time/effort necessary to recognize audits (so as to be sure to pass them)? I believe that the only thing that your observation demonstrates is that, unlike the other five categories, the suggested edit audits are far too obvious, to the point of being trivial to recognize quickly by even a robo-reviewer, and so we need to make them trickier. And you realize, of course, that I'm not proposing removing the cap, just loosening it a little. ... (Cont'd) Sep 7, 2016 at 20:02
  • (Cont'd) ... But this raises another point: it's my opinion that fewer than 66% of suggested edits are worthy. We may quibble over the number, but I expect that most of us will agree that there are an awful lot of awful suggested edits. I've been thinking for a while that anybody who approves more than N % of suggested edits should be looked at closely by the moderators (i.e., such behavior should set off an alarm). N might be between 80 and 90. Then again, the robo-reviewers would adapt to that by randomly rejecting some suggestions, without looking at them. Sep 7, 2016 at 20:03
  • @Scott As long as there's a cap, things should be OK. (We can discuss the number of the cap when we get there). Regarding edit audits, I've long had a request to make them harder: Generate new types of audits for the Suggested Edits queue . Sep 8, 2016 at 6:41
  • 1
    @Scott Regarding Approve/Reject rate, I find that an Approve/Reject rate of 5:1 or higher is a red flag. Not proof, but definitely an indication that something could be seriously bad. Tangentially related: I once asked Shog9 in chat if they also look at the speed with which users review. He said they do. But that's all I know. Sep 8, 2016 at 6:47
  • 1
    @S.L.Barth: So you suggest N = 83.33 — sounds reasonable. I was also thinking that speed would be a useful metric; it's good to see that some people are thinking along the same lines I am. :-) Sep 8, 2016 at 15:17

You must log in to answer this question.

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