What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 133 Stack Exchange communities.

Now that I can review suggested edits, I've noticed that occasionally, the edit comments don't match the changes made to the post.

For example, in this post, the comment for version 3 is Fixed title and some typos but an entire paragraph was deleted!

Obviously, it's hard to tell whether it's in bad faith or just accidental (probably the latter).

Should edits be rejected when there's a glaring mismatch? Should the proposer be notified by the reviewer that the comment was misleading?


Update: apparently that was a bad example, because it was caused by SO's handling of concurrent editing and not by any user. I do not have any good examples at the moment.

share|improve this question
7  
Some more examples ;-) –  Arjan Nov 30 '11 at 20:20
1  
Standing out there, in the pouring rain, you can taste the feathers and dust of disappointment scar and stain the back of the throat. A burning sensation crawls up from the side leg, prickly with anticipation and desire. Quickly then to fall over in a heap of despair, delusion and dirge. -- wow no kidding. –  Matt Fenwick Nov 30 '11 at 20:24
    
Since the edit comments cannot, themselves, be edited, there was some consternation a while ago about people leaving offensive messages in them. See Allow mods to edit revision comments. –  Pops Nov 30 '11 at 20:31

2 Answers 2

up vote 6 down vote accepted

What happened there was the editor started editing based on revision 1. The asker edited that question to add that "paragraph" before the editor finished and saved revision 2. The editor finally finished editing and submitted the change. The asker wasn't done editing the question, saw the suggestion (or most likely not) and "improved" the suggestion. The suggestion was implicitly approved when the asker finished the edit finishing with revision 4. Later on, the asker realized that the previously added "paragraph" was removed and added it back in to revision 5.

It's not the fault of either party, it's just the current system doesn't handle this situation very well.

This is exactly the kind of scenario that I want stopped or avoided. There should always be confirmation if your edit will step on top of another recent edit. Right now, all "significant" edits will go through with no notification. The removal of that "paragraph" was considered significant and therefore went through with no problem.


But to answer your question, you shouldn't rely solely on the comment being misleading as the deciding factor, you should be more concerned about the changes it made to the post. It ultimately should improve the post so it will be easier to read for everyone. As long as the comment isn't downright offensive or is useful, I see no reason why it should be rejected because of it. Look at the comment as the editor's justification for the edit, not a complete summary of changes.

share|improve this answer
    
Oh wow, so what I ascribed to carelessness was actually due to a bug/feature of SO. Any idea what the status on that is? No answers on that last question. –  Matt Fenwick Nov 30 '11 at 21:20
    
There hasn't been any progress with my request AFAIK. I even placed a bounty on it a couple of weeks ago. Got a few new viewers but nothing said by the team. –  Jeff Mercado Nov 30 '11 at 21:23

If an edit removes an entire paragraph, it could probably be rejected as too radical, or approved because it is removing noise. To make an example of noise, suppose that the user who asks a question starts explaining that he asked the question on a different SE site, and that question was closed as off-topic, after he was suggested to ask it on a different SE site; then the user continues saying that he doesn't agree with that decision, but he is forced to ask the question in an SE site that he doesn't repute to be the best one for his question.

Suggested edits should be rejected if they are not valid, not because the comment is only partially describing what the edit is doing. If you don't understand why the edit is being proposed, and you are not able to understand if it is a valid one, you should neither approve it nor reject it.

share|improve this answer
    
Agreed that they shouldn't be rejected as long as they're valid. But the example where a paragraph is removed -- regardless of why it's removed, the reason should be mentioned in the comment. –  Matt Fenwick Nov 30 '11 at 20:44
    
Even if the reason would have been mentioned in the comment, the proposed edit was invalid, IMO. The comment helps in understanding what is being done and for what reason, but it doesn't make the proposed edit valid. –  kiamlaluno Nov 30 '11 at 20:48

You must log in to answer this question.

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