[TL;DR? Skip to "Bottom Line:" at the end. But I hope I made my post worth the length]

I was editing a post from the review queue. Apparently, if another edit is submitted while you're editing from a review, you don't get the prominent warning you get in non-review edits, with the conspicuous orange bar at the top of the page. You just get a very subtle notice in black-on-white text that's easy to miss (I only saw it when backtracking through the browser cache). So, the result was this: https://stackoverflow.com/review/suggested-edits/2646821

As you indicated by my comment, I had actually made substantial formatting improvements to some very poorly formatted code, but in comparison to the previous edit, very little was left. But what's far worse is what happened above the code. As my comment said, I considered the question itself garbled beyond repair, and was leaving it to the OP to figure out what he wanted to say and then try to say it in English. However, the previous editor had made a valiant attempt at tackling that task, so my edit had the effect of restoring the worst parts of the original post.

I wanted to cancel my edit immediately, or withdraw it, or whatever you want to call it - roll it back, axe it, incinerate it, bury it and salt the ground above it, wipe it from the face of the Earth and from all records and all memory and make it illegal to mention that it had ever existed, like the an act of the Ministry of Truth in 1984. But I couldn't.

Sure, the edit was rejected, as it should have been. But three editors wasted their time on it, which they shouldn't have. And it bugs me to no end that this edit made me look like a complete idiot. The reason given by two of three editors was "This edit changes too much in the original post; the original meaning or intent of the post would be lost." Why? Because the side-by-side comparison makes it look like I actually added "Hi please kindly clear my doubt I have shown" and "Pls click edit given below button showing full coding" (ironic considering this other edit I had just made - see my comment: https://stackoverflow.com/review/suggested-edits/2646434), and even changed "and" to "adn"! It doesn't even give any indication that my edit is being compared to another edit; it looks as if my edit is being compared to the original. Heck, based on the information as it was presented to them, I'm surprised that none of the editors selected "vandalism" as the reason for rejection!

Note that this question has been asked before - Cancel a suggested edit - but it was marked as a duplicate of this post: Can I modify and resubmit my suggested edit before it is peer reviewed? I strongly disagree that it was a duplicate. The question in the earlier post addressed a different issue, and the answer given was relevant to that issue and not this one. There's a big difference between wanting to resubmit an edit in order to improve it, and wanting to just outright withdraw an edit because it conflicts with another edit that it "crossed in the mail" (and possibly makes you look ridiculous to boot). The "grace period" to modify your edit while it's pending review doesn't help much if you just want to defer to the other edit that was submitted while you were working on your own edit.

My question is a duplicate of the falsely accused duplicate, but unfortunately that post was prevented from getting discussed by being marked as such.

[It's a separate topic for another time, but one thing I see all too frequently at SO is that some people are way too hasty to label anything that kind of sounds similar on a superficial level to something previously posted as a "duplicate". Here's an example of someone flagging one of my SO posts as a dupe, which completely missed the entire point of my question: What's the difference between the "throw" keyword and "Write-Error -ErrorAction Stop"? The question I was allegedly duplicating may sound similar, but it wasn't even germane to what I was asking].

Bottom line: How about adding a feature to cancel an edit you suggested while it's pending review, so that if you realize you made a mistake or missed an update and are rolling back some improvements rather than making improvements, you can just yank it out of the queue so it won't waste the reviewers' time and in some cases make you look like a total jackass when you actually had made a good edit? Also, how about making the warning you receive about other edits being made while you're reviewing as conspicuous as the usual warning you get if another edit is made while you're working on an edit?

  • tl;dr. But, on SO the suggestion queue rolls so fast that you won't have the time to retract your suggestion anyways. Aug 3, 2013 at 9:05
  • 2
    Granted, this question is quite well-written and well-justified, so you deserve all the upvotes you can get off of it. But the fact still remains that you're not the first person to have this idea. :-)
    – Cody Gray
    Aug 3, 2013 at 9:07
  • did read; thanks for the amusing post. Particularly, the third paragraph. Aug 3, 2013 at 9:08
  • Pretty much what has been said... excellent post, but unfortunately a duplicate many times over. In the end, it would be a nice feature to have but (at least on SO) there is almost zero value as you have to know that you want to cancel the edit immediately to catch it before the reviewers finish reviewing. Aug 3, 2013 at 9:14
  • To summarize the scenario: You were suggesting an edit on revision N of a post, but another >2k user has already edited the same post (creating revision N+1) before you submitted the suggested edit. So when reviewers see your edit, it's compared to revision N+1 instead of revision N, so it seems like you're vandalizing the post when you're actually not. Aug 3, 2013 at 9:16
  • @CodyGray But as I explained in my post, I think that other question addresses a fundamentally different issue. There's a difference between wanting to cancel in order to resubmit an improved edit (which is adequately addressed by the answer given to that question - you don't need to cancel, you can just edit your edit while it's under review) and wanting to simply withdraw your edit altogether (which is not addressed at all by that same answer). I know I'm not the first person to have this idea, but the last person had his question closed as a dupe of that fundamentally different question. :)
    – Adi Inbar
    Aug 3, 2013 at 9:19
  • @psubsee2003 See the previous post that was marked as a dupe of the earlier post that addresses a different issue - that combined with my experience demonstrates that it's not unlikely that you'll know you want to cancel immediately after submitting. That's when you see that someone else edited the post, and you're unintentionally editing some of their improvements back to the original version. Or in some cases your edit is very close to the previous edit, but then, why have people waste time on it instead of letting the editor just withdraw it?
    – Adi Inbar
    Aug 3, 2013 at 9:24
  • @AdiInbar the 2nd one in the list is actually a duplicate, not the first. Covers exactly the same ground in a less humorous way. I was going to that one as a dup of this one though since it is higher quality post Aug 3, 2013 at 9:27
  • 1
    @OldCheckmark Yes - but even worse, there's no indication to the reviewers that there even was a revision N+1! The OP's name appears on the left side, so some things that were rightly taken out in revision N+1 look like things that I composed and added to the post for no reason!
    – Adi Inbar
    Aug 3, 2013 at 9:27
  • @JanDvorak Actually, if I could have whacked it when I first wanted to, it wouldn't have even reached the first reviewer. I spent the next 10 minutes or so looking for a way to undo my edit, but the only "undo" available was to roll back from the previous edit to the original, which was even worse...far, far worse.
    – Adi Inbar
    Aug 3, 2013 at 10:27
  • Excellent question - though you might want to request a more prominent "an edit has happened" indication instead. Or let the review show the diff to the state your suggestion is based on, and then attempt some git rebase-like merging Aug 7, 2013 at 7:30


Browse other questions tagged .