What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 127 Stack Exchange communities.

This question already has an answer here:

I just can't fathom how terrible edits like this get through - all three reviewers approved. All he did was bold (instead of block-quote) the error message, and change one barely comprehensible sentence into something equally incomprehensible, leaving lots of other minor problems there (which I have subsequently fixed).

Do we need to introduce some kind of penalty for suggesting terrible edits, and to the approvers who blindly let them pass? Obviously the audits aren't keeping people honest.

This is probably a duplicate, but nothing obvious popped up immediately. I saw this question but it seems to deal with the opposite case - most of the problems were fixed but not all. It is also talking about the situation where a user is trying to improve an edit and it is approved right out from under them.

share|improve this question

marked as duplicate by Servy, Hugo Dozois, Aaron Bertrand, Shadow Wizard, Martijn Pieters Oct 25 '13 at 20:18

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.

4  
    
@Al thanks, will review those. –  Aaron Bertrand Oct 25 '13 at 17:36
7  
Based on past experience and an ungodly number of meta questions on the subject (see Al's comment just to get started) the only solution I've found to keep my sanity is to stop reviewing suggested edits very often. It would seem to be an unsolvable problem. –  Servy Oct 25 '13 at 17:37
4  
You only think it's a bad edit because it didn't fix everything. –  Robert Harvey Oct 25 '13 at 17:38
    
Posting to meta and hoping that the reviewers that so obviously deserve a manual review ban get noticed by the mods is certainly appealing :) (Not that this is a particularly egregious example, but one of the reviewers also approved changing "i" to "I" in a post very recently....) –  Wooble Oct 25 '13 at 17:38
2  
Even if the edit did not make the post perfect, it is still an improvement over the original. At least the error message is visible. I would have also approved it (if I was too lazy to improve it - that is) –  user000001 Oct 25 '13 at 17:38
6  
Given that, if I would have improved it, I would have undone half of what the editor added, in addition to making quite a few changes that were missed, I would most certainly not have marked it as helpful. –  Servy Oct 25 '13 at 17:40
7  
@Robert (1) personally I don't feel it fixed anything. The added word in the sentence did not improve readability and I don't find a bolded error message easier to read (the only thing I'll concede is perhaps the carriage return) (2) the problem with accept and improve is that it can get approved right out from under you. And my issue is that these three reviewers just clicked "all good" without bothering to improve something that was obviously in need of further improvement. So my problem is more with the blind reviewing than the original, sub-par, suggestion. –  Aaron Bertrand Oct 25 '13 at 17:43
    
@Servy this is true, but I am wondering about what can be done about terrible reviews after the fact. This is clearly not a suggested edit I can act on and improve, because it has been approved long before I ever came across it. –  Aaron Bertrand Oct 25 '13 at 17:45
3  
@AaronBertrand See my first comment. The only solution I have found is to give up (and that's after a lot of time trying to find a good solution). If I find a better one, I'll gladly share it. –  Servy Oct 25 '13 at 17:46
1  
@Servy Ok, so four votes for "give up." I'm still holding out hope that something more constructive can come from the discussion. :-) –  Aaron Bertrand Oct 25 '13 at 17:47
8  
I would love to be able to either flag a review or a reviewer for moderation attention. This has been suggested before so I don't hold out for a lot of hope on this. –  Some Helpful Commenter Oct 25 '13 at 17:56
    
There should be points deducted from anyone who passes a first post that later gets flagged, or accepts an edit that later gets flagged. –  Johnny Bones Oct 25 '13 at 18:46
2  
@JohnnyBones This meta question has nothing to do with the first post queue. While there is some overlap between review queues, dealing with bad reviewers in different queues is still quite a bit different. Suggested edits, especially, is quite a bit different from all of the other queues. –  Servy Oct 25 '13 at 18:50
    
@AaronBertrand: Fair enough. –  Robert Harvey Oct 25 '13 at 20:09
show 1 more comment

1 Answer

I don't think we should punish the person who suggests such an edit (unless in very extreme cases where the edit is obviously mean spirited, so in cases the vandalism reject reason applies). The reason suggested edits by users with less than 2000 reputation go into review is that they have not yet been around for very long, and thus may not yet have picked up the expected quality standards. It's learning by trial and error: you edit something and it gets rejected, you look at why it was rejected and try to improve in the future. In the very beginning I, for some reason, held the opinion, that I should completely rewrite questions that I found unclear to better match "the OPs intentions". These reviews got rejected, and rightfully so, and I adjusted to that.

TL;DR: Don't punish the person suggesting a bad edit.

Now, on the other hand, we let people review that do have enough experience on their backs to know what kind of edits improve a question, and which should be rejected. For the whole trial and error system to work, we need reviewers to pay attention, and accepting edits that clearly should not have been accepted damages the system, and confuses the inexperienced users. Now sometimes edits are rejected, but sometimes they are accepted. You cannot learn from that. That's also why it takes multiple votes for a suggested edit to be approved (or rejected for that matter), because a falsely accepted edit actually causes harm beyond the fact that the content edited was not improved or even worsened.

TL;DR: Persecute and stake bad reviewers.

share|improve this answer
1  
Yes, I have a much bigger problem with the blind robo-reviewers just clicking go and next. However, I am also concerned that they are indirectly giving the suggesters feedback that their suggested edits are good, and I think that's a related but different problem. –  Aaron Bertrand Oct 25 '13 at 18:56
    
But in my opinion this is, also, caused by the robo-reviewers. Kick those out and people will stop suggesting bad edits, since there is nothing to gain from that. –  nijansen Oct 25 '13 at 18:57
3  
I was going to vote, but TL;DR: It's okay to suggest bad edits. That's a -1 from me. If there's no bad edits, there are no bad edits to review. TL;DR: I think we should persecute and stake bad reviewers. That's a +1 from me. Net gain: zero! –  LittleBobbyTables Oct 25 '13 at 18:59
    
@LBT I can live with that. As in any discussion there is no right or wrong, just agreement or disagreement. Oh, and rhetoric. –  nijansen Oct 25 '13 at 19:02
1  
I would say that a better phrasing is that "it's expected that there are going to be bad edits". One can never ensure that bad edits are never made. However, one should ensure that bad edits are never (or at least very rarely) approved. That is, however, currently not the case. In any case, I don't see Aaron as suggesting that editors be punished, but rather that the problem is with the reviewers. –  Servy Oct 25 '13 at 19:04
2  
@Servy - i.imgflip.com/4ebfn.jpg –  LittleBobbyTables Oct 25 '13 at 19:06
    
@Servy I totally agree with that, also that my phrasing may not be very good for my TL;DR part. –  nijansen Oct 25 '13 at 19:06
    
if reviewers do their job right and reject crappy suggestions then person who suggests such an edit eventually gets edit-ban if they don't learn from rejections –  gnat Oct 25 '13 at 19:07
    
@LBT and a natural +1 for a hyperbole and a half image ;) –  nijansen Oct 25 '13 at 19:07
add comment

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