What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 134 Stack Exchange communities.

This question appears to have had an edit collision between revisions 3 and 4. Two different people have edited it with the same tags. This has resulted in creation of a new revision that effectively serves no purpose.

This is a problem because:

  • the question has been unfairly pushed closer to community wiki status

  • it's unclear what happened from the revision list page

Is it possible to ignore subsequent revisions that make no effective changes (at least in regards to tagging)?

share|improve this question
    
    
Can't find dupe, please add as comment –  Alex Angas Sep 10 '09 at 11:10
1  
For the image alone: meta.stackexchange.com/questions/20658/… There are others, but this one can has cats. –  random Sep 10 '09 at 11:18
    
@random: I see they're related but still this is distinct. Thx for the lolcat –  Alex Angas Sep 10 '09 at 11:20
    
possible duplicate of Preventing conflicting simultaneous edits –  Gnome May 30 '10 at 7:29

1 Answer 1

up vote 1 down vote accepted

In the general scenario, if two edits are identical in content then the second edit (henceforth referred to as "faux edit") is ignored. It is as if nothing was submitted at all. The edit summary doesn't matter. So in the general scenario, identical revision changes are ignored as you request.

This is independent of both the 5 minute edit window and who is actually editing - if I do a faux edit on one of your posts three days after you last edited it, it will be just the same as if I do a faux edit to one of my posts 5 seconds after I realized I had a typo in my edit summary. This latter point means that if you do want to change your edit summary in the 5 minute window but don't have any other revision to make, you need to perform 2 really fast edits that cancel each other out.

Humorously, you can get caught by the CAPTCHA for a faux edit, but even passing validation you won't actually submit anything material and the post will not be modified in the slightest.


The Bug - What Happened

If, while one person has opened the edit window for a question, another person completes an edit (be it through the 10kr inline tag editor, 500r retag tool, or just plain editing the question) that is identical in content, the first person's edit does not count as a faux edit. It doesn't matter whether the edit is for tags or post body.

This was replicated as part of the test question linked from here by myself opening the edit window, then leaving it open until Jon Seigel made his edit. When he made the edit, I opened a new edit window in a separate tab, copied the post body from Jon's revision, and pasted it back in the first edit window and submitted that. And, just like magic, we ended up with a new revision that provides absolutely nothing new. Except maybe an edit summary, but as the question asker's example shows this is unnecessary.

Why does this happen? The most likely scenario is that the identical content check is not made on the actual post body, but on whether the text box on the form is changed. Which kinda makes sense. As such, since the new post is different than the cached post body, it will submit the change regardless of the fact that the new post is identical to the actual post body.

Additional note: According to Jon Seigel, this collision will also prevent the first editor from being able to rollback the change.

share|improve this answer

You must log in to answer this question.

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