5
$\begingroup$

This is somewhat related to this meta post which received no clear answer.

I recently tried to edit the accepted answer here by adding crucial information its writer wrote in the comments, but that was otherwise missing (this is a link to my suggested edit).

In my opinion, the original answer without the answerer's comment is somewhat unclear, and adding the additional information would contribute a lot to its clarity and usefulness. However, my edit was rejected for changing the original answer too much (in particular, changing its intent, which I don't think I did).

What should I do in such a case? Should the edit have been accepted? And if not, is it legitimate to write a new answer to the question that combines the other answer + the comments and together makes a better (in my opinion) answer than the currently accepted one?

$\endgroup$
  • 5
    $\begingroup$ Additional possibilities: (i) an answer describing itself as supplementary information to the first answer; (ii) add the comments at the end of a question Eg "[Edit by Dean Gurvitz: ___ gave additional information in comnents, which I've collected together here . . . ]" So you're not changing anything in the answer, just adding to it. $\endgroup$ – timtfj Feb 4 at 20:52
7
$\begingroup$

Editing answers to improve them is a generally good idea, but sometimes people don't like it when posts are edited away from the OP's intent. There is a wide gray area in the middle that's a bit hard to navigate.

But should you find yourself uncertain, it is always find to post a better, complete answer. Further, one should think of comments as being fundamentally temporary --- sometimes they get deleted and no one really thinks twice. When asking yourself "Hmm, should I collect a complete answer from a collection of comments", the general answer is yes, go for it.


There are also semantic reasons for making a complete answer, especially as opposed to a series of comments. People vote and edit posts, not comments --- so small typos in comments are more challenging to note and correct (especially within long comment threads).

Retrieving posts through the StackExchange API makes piecing together comments into a single very annoying. Similarly, having piecemeal answers affects search engines when they crawl the site --- but having a single complete answer makes it more search-visible.

And still further, navigating comments happens to be much more annoying through accessibility software (such as screen readers), to the extent where I've seen many people using accessibility software simply ignore comments altogether.

$\endgroup$
  • $\begingroup$ Thank you for your answer. The thing that really bothered me in this specific case was that my edit really wasn't changing the answer "away from the OP's intent", and still got rejected. $\endgroup$ – Dean Gurvitz Feb 4 at 20:00
  • 2
    $\begingroup$ @Dean In your comment on the edit, did you make clear that you weren't just adding information from comments but from the answerer's comments? The reviewers might not have realised. $\endgroup$ – timtfj Feb 4 at 20:42
  • $\begingroup$ I don't think I did, but a quick glance at the answer's comments would have shown that the only comments there are questions from the original asker and then replies from the answerer. It still can provide an explanation as to why the edit was rejected. $\endgroup$ – Dean Gurvitz Feb 4 at 20:45
  • 2
    $\begingroup$ @Dean I've just become able to review edits, and the way they're presented for review doesn't make it easy to get at the comments to read them. I think that might be part of the problem. Also it doesn't seem possible to see answers other than the one you're reviewing. $\endgroup$ – timtfj Feb 4 at 23:50
  • 2
    $\begingroup$ "always find" should be "always fine"? Yes, I was tempted to edit myself... $\endgroup$ – hardmath Feb 5 at 2:27
  • $\begingroup$ @timtfj Maybe you can use this opportunity to suggest new features/capabilities to reviewing edits! $\endgroup$ – Dean Gurvitz Feb 5 at 6:30
6
$\begingroup$

Nice edit! I can see why you'd suggest that edit, and why you'd be confused that it was rejected.

Let me share a little background that might make this a bit more comprehensible. This site appears to have a culture of rejecting major edits to other people's answers, presumably on the basis that you are putting words in someone else's mouth or deviating from their style. (Personally, this wouldn't be my preferred policy, and I realize it's a bit confusing given what our help center says, but it is what it is. And I can appreciate some potential reasons for that position; e.g., it's hard to review a suggested edit for correctness, when making a substantial change like this.) I don't know if this a hard-and-fast policy or just a common opinion among the folks who tend to review edits, and it doesn't always happen that way, but it's a tendency I've noticed in my personal experience. So, while I think that your work is laudable and the edit you prepared would have benefited the site, be prepared that such edits are often rejected.

What should you do instead? I've come to the conclusion that the most pragmatic approach is to post a new answer. Take the information in the existing answer plus the comments you found helpful, and post your own answer. Make your answer complete so it can stand on its own. Acknowledge your sources for any copied material, and credit the people who gave you useful ideas. This way, we still get a good clean answer with a nice self-contained explanation of a solution.

I hope you won't be discouraged. It's not that your work is bad. It just doesn't quite fit into the established practices and norms here, so my advice is to work around those practices and find another way to post the good answer you put together. Thank you for working to improve the site and the quality of answers here!

$\endgroup$
  • $\begingroup$ I understand and am not discouraged! Eventually what I did following davidlowryduda's answer was exactly to write a new answer to the original question mentioning where I got the ideas from. $\endgroup$ – Dean Gurvitz Feb 5 at 6:31

You must log in to answer this question.

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