105

A user recently suggested a great edit which added useful information to one of my answers. However, because it was a substantial edit, it was rejected by the reviewers. That's fine. However, as the owner of the post I would like to be able to overrule their decision after reviewing the edit myself (as I was encouraged to do by the notification in my inbox).

Currently, I had to open the edit suggestion in a different tab and manually copy and paste over the changes from the markdown diff to apply the changes. This seems laborious and doesn't give credit to the user who suggested the edit.

Request:

I would like a button on the suggested edit page, visible to the owner of the post, which applies the suggested changes, preferably just as if it had been accepted by the reviewers (i.e. giving credit to the editor and the +2 reputation for editing, if applicable).

Related discussions: (note that since this is a feature request, these are not duplicates)

Improve this question
10
  • 3
    Something that isn't mentioned here that I brought up in mu duplicate question is that overridden reviews could provide meta-data on the reliability of reviewers.
    – Sam Hasler
    Mar 21, 2014 at 13:54
  • 4
    @SamHasler not really. Often the rejections may be in accordance with the reviewing guidelines of the site, but the OP still thinks the edit was helpful. How would this be a fair judge of the reviewers?
    – psubsee2003
    Mar 21, 2014 at 13:56
  • 4
    @psubsee2003 if a reviewer had their reviews overruled more often than other reviewers it could be a signal that they are not doing them well, in which case you might not want to allow them to review as much as someone that does not.
    – Sam Hasler
    Mar 21, 2014 at 14:55
  • I propose an alternative: give the author of a post 24 hours to approve or reject the edit before it enters the review queue.
    – Harry Johnston
    Sep 16, 2014 at 4:04
  • 1
    Or, perhaps, provide a way for the editor to say "this is something I'd like the original author to review" ? (That might be useful even for those with edit privilege.) Sure, most of the time you can propose such a change in the comments, but in some cases (code changes in particular, or adding sample code to an answer) that isn't feasible.
    – Harry Johnston
    Sep 16, 2014 at 4:06
  • 2
    I wonder why this never become implemented?
    – Trilarion
    Jan 14, 2015 at 16:38
  • 3
    What ever happened to this feature request?
    – Praxis Ashelin
    Feb 27, 2015 at 15:39
  • 2
    I just have been bitten by this too and need to ask the same question.
    – martin
    Apr 23, 2015 at 12:37
  • 1
    Almost the same request here: meta.stackexchange.com/questions/178340/…
    – serv-inc
    Feb 2, 2016 at 9:32
  • So, how can we revive this ?
    – Denys Séguret
    Aug 6, 2016 at 7:06

5 Answers 5

Reset to default
44
+100

Where the post hasn't been altered since the attempted edit this seems like a fair proposal, given that the OP can simply edit it in anyway giving credit seems fair. Presumably with a message along the lines of;

enter image description here

However if another edit has since been made to the post it would become logistically impossible; should the OP choose between the two, attempt to combine them? I think at the point where another edit has been successfully made the rejection should become permanent.

Improve this answer
7
  • 1
    Well the OP could still have the chance to mark the edit as "helpful"/approved even if the incorporation into the answer is a manual process.
    – Martin Smith
    Sep 26, 2013 at 14:01
  • @MartinSmith While I like that as an idea it could be confusing for a user to approve an edit (to their mind) and yet nothing happen. Maybe a next step if this is approved?
    – Richard Tingle
    Sep 26, 2013 at 14:11
  • 2
    We need a merge in the system :) Someone let's get the fishing poles ready, 'tis a can of worms we'd open.
    – Kuba hasn't forgotten Monica
    Sep 26, 2013 at 16:44
  • 5
    I retrospectively approve this suggestion.
    – jmac
    Oct 1, 2013 at 2:01
  • 5
    I think this is the best way to handle this, but I would not support the reject button. The OP should have no business retroactively rejecting suggested edits because more often than not they are rejecting it for the wrong reasons (i.e. don't like other people editing their question). I have a feeling it would lead to more "rollback wars".
    – psubsee2003
    Oct 5, 2013 at 9:01
  • 4
    @psubsee This dialogue would be only for already rejected things. My reject button is more "keep rejected"
    – Richard Tingle
    Oct 5, 2013 at 9:20
  • I don't know that I see the need for the "keep rejected" button, but I'm very glad to see "given that the OP can simply edit it in anyway giving credit". I've been reading through some of the related questions, and the general advice is to "just make the edit", but that seems like the suggesting editor could quite legitimately flag for plagiarism afterward. Having an "Approve" button could preserve the expected edit history, and ensure attribution where it should be.
    – Joshua Taylor
    Feb 1, 2014 at 15:23
15

As far as letting the owner of the post retro-actively approve already rejected edits, I sense some logistical problems with that.

BUT It does make sense to allow the owner's single vote to completely approve/reject the edit; giving them final say over the review process.

IF an edit that they liked gets rejected, there's nothing stopping them from going back and making it them self, and you still don't have the logistical problem of rolling back to edits that never actually got applied in the first place.

Improve this answer
11
  • 7
    "BUT It does make sense to allow the owner's single vote to completely approve/reject the edit; giving them final say over the review process." That's already the case.
    – Servy
    Dec 6, 2012 at 17:29
  • @Servy in that case, the system works at it should.
    – Sam I am says Reinstate Monica
    Dec 6, 2012 at 17:32
  • 12
    @Servy: Yes, I believe the owner is currently able to approve with a single vote, but only if it hasn't already been rejected. With the speed at which edits get reviewed in the new review system, that's likely to happen before the owner has a chance to even look at it.
    – hammar
    Dec 6, 2012 at 17:44
  • @hammar Yes, I'm aware of that, which is why I specifically quoted Sam. Clearly it's not particularly relevant to answering your question. I was simply informing Sam of something he apparently didn't know.
    – Servy
    Dec 6, 2012 at 17:50
  • 3
    @SamIam Re: "there's nothing stopping them from going back and making it themself" If the owner makes the edit and doesn't properly attribute the content to the suggester, then couldn't the suggester legitimately flag the post as plagiarism (with a link to the suggested edit)? I think the ability to "accept suggested edit despite rejection" would address this issue because the suggestion, etc., would then be present in the edit history, etc.
    – Joshua Taylor
    Feb 1, 2014 at 15:21
  • @JoshuaTaylor no, the suggester couldn't flag the post as plagiarism because it's not plagiarism.
    – Sam I am says Reinstate Monica
    Feb 1, 2014 at 20:40
  • @SamIam Answers get deleted all the time because they copy text from other SO contributions without attribution. According to the footer of the site, "user contributions licensed under cc by-sa 3.0 with attribution required". Attribution is typically provided by the edit history; if the owner of the post copies and pastes text from a rejected suggested edit without attributing the text to the suggester somehow, it seems that the content is being used in violation of the under which the suggester contributed it. Why would this be different from any other case where the author of an answer…
    – Joshua Taylor
    Feb 1, 2014 at 20:49
  • …copies text from somewhere else and pastes it into the answer without attribution?
    – Joshua Taylor
    Feb 1, 2014 at 20:50
  • @JoshuaTaylor But the suggesters are not creating new text; they're modifying yours
    – Sam I am says Reinstate Monica
    Feb 1, 2014 at 21:10
  • @SamIam That's a good point, and probably sufficient for most edits. Not all edits are small though; I started searching MSO for this issue because of a rejected suggested edit to one of my answers. It added a new paragraph to my answer, and while I didn't end up copying it in verbatim, it was much more significant than minor typo fixes or things of nature. If I had simply copied and pasted the content, I don't think it would have been unreasonable for the answer to get flagged for unattributed content.
    – Joshua Taylor
    Feb 1, 2014 at 21:16
  • @JoshuaTaylor no moderator would uphold that flag
    – Sam I am says Reinstate Monica
    Feb 1, 2014 at 21:21
9

I had an edit to my post that was made by the user asking the question, who tried out the code, debugged it for me, and posted the bug fixes.

Unfortunately, it was hastily rejected by 3 different users. In order to fix my post I had to study the diffs and manually apply them.

I have some proposed solutions:

  1. Even after an edit has been rejected, show an "Improve" button for users with privileges to edit the post without review (i.e. the original author or an high-rep user). This may only apply if there has not been a later edit.

  2. Have a link to see the source, allowing me to copy and paste the whole rejected post into the edit box on actual post. It would also be useful to see the source on the original post so I could compare them with my own diff utility (this would be handy when SO's own diff display is insufficient).

Improve this answer
2
  • He could have said he was the OP and that he was fixing a typo, instead of writing this as the comment: some changes i made in your code which will avoid errors :)
    – dcaswell
    Sep 30, 2013 at 13:35
  • 4
    @user814064: It's pretty clear that the OP doesn't speak English that well, and it's unlikely that he would a priori the magic keyword to get his edits approved. The box says "Edit Summary", not "Tell us why we should approve your edits".
    – Gabe
    Sep 30, 2013 at 16:41
2

Another design possibility would be to put it close to the Next button:

enter image description here

The code to do that should already be in the system. So it would be just about adding another button, which says f.ex. Retroactively approve.

Improve this answer
-7

Even if the author of a post would like an edit to be applied in cases like this, the fact remains that the person who suggested the edit was violating the rules for editing, and it's important that the edit be rejected for that reason. If the edit would be retroactively approved by the OP, even though it was correctly rejected by reviewers, then it gives the editor the impression that such an edit is a good edit, and that they should continue making edits like that in the future. That is not the case. It is a violation of the system for them to continue making such edits, and will just result in the reviewers needing to reject more of them in the future.

It's important for such an editor to learn the proper way of adding such content to the site, namely that this content would be best posted as a comment to your answer, or in the event that the edit has sufficient useful information, even its own answer. If the author of the post acknowledges that they want the content in their answer then they can add it themselves, or give permission for the user to edit it in (at which point it would be an appropriate edit). If edits such as the one you linked to are approved then users won't learn the proper workflow for such content.

Improve this answer
8
  • 6
    Often they aren't violating rules for editing. If an answer is generally on the right track but the OP finds when implementing that it needs some changes to be a complete answer these edits Improve the post and should be fine. Reviewers should skip them if unsure. Not reject.
    – Martin Smith
    Sep 26, 2013 at 15:32
  • 2
    @MartinSmith If they're adding content then that's not what the edit system is there for. If someone has new original content that they would like to contribute to a q/a then they should either comment or add a new answer, not just pick some random answer and edit all of their own contributions into that one.
    – Servy
    Sep 26, 2013 at 15:41
  • 1
    Here's another example. The OP submitted a suggested edit to correct an inaccuracy. Reviewers rejected it. The answerer manually applied it. I agree with Gilles's answer there and Shog's here.
    – Martin Smith
    Sep 26, 2013 at 16:09
  • @MartinSmith That is an entirely different case. It's fixing a minor mistake in the post, whereas this is adding new, original content that's not directly tied to the answer itself, or any mistake within it. "minor" is also rather key there. If the mistake is a core issue with the entire post, and fixing that mistake would be a major re-write, then editing is not appropriate, but when it's a small change, like the example in that link, then editing can be appropriate.
    – Servy
    Sep 26, 2013 at 16:10
  • 1
    This feature request would apply to that case too.
    – Martin Smith
    Sep 26, 2013 at 16:12
  • @MartinSmith Now you're describing a case where the reviewers rejected an edit that they shouldn't have rejected, but the OP wants to approve it, which is radically different from the cases where the reviewers correctly rejected an edit and then the OP wants to approve it. The issue in that case is that the reviewers should have approved the edit. As to this feature request, it comes down to who is more likely to know which edits should be approved, the reviewers or the author. My (rather considerable) experience would indicate that authors are not likely to be correct more often.
    – Servy
    Sep 26, 2013 at 16:14
  • 3
    The issue isn't black or white though. One man's radical edit is another's improvement, I think often reviewers will simply reject code changes out right even if not qualified to judge. I would vote to approve if they are correct and improve the answer or skip if unsure.
    – Martin Smith
    Sep 26, 2013 at 16:23
  • @MartinSmith Reviewers rejecting code changes that are appropraite is an issue that you're welcome to discuss, but the solution should really be finding ways to ensure that reviewers approve them when they should, not finding ways for authors to come back and revert the review decision, simply because the reviewers are more likely to be right than the author. The author can still add the content if they wish, they just can't indicate that the editor was acting correctly, because it's likely that they weren't.
    – Servy
    Sep 26, 2013 at 16:28

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