How invasive of an edit is too invasive?

Question

I recently made an edit to an answer and another user raised the question of whether it was "kosher." In this case, I feel pretty confident that it's reasonable (or else I wouldn't have done it). On that question, there was a comment that contained a pertinent fact (who gave the "Moonlight Sonata" its moniker). The author of the accepted answer made reference to this comment but didn't duplicate the information. However, since someone was recently pointed to this question and said that it didn't contain the answer they were looking for, I felt it was reasonable to spell out in the accepted answer the data that the author was referencing.

I'll bring up a slightly different situation I've thought about recently (on a different SE). A user asked a question that felt like it ought to be a duplicate. A search revealed an older question that was a near duplicate, but brought in a shade of emphasis that wasn't in the new questioner's intent, or vice versa. I considered encouraging the questioner to edit the old question to allow it to serve both their purposes, but ultimately felt it would be too invasive. If I recall, it wouldn't even have violated the original intent of the old-questioner, and their questions were really the same, but the edit would have removed elements that were referenced in comments or answers, so I ultimately bit my tongue.

Answer 1

I have a vague and unhelpful answer, and I imagine it's the only real answer to this question, but just floating it to get others' perspectives.

I imagine the answer is "An edit is too invasive when it violates the intent of the author." When you present them as saying something they didn't mean to say. I'd say it's too invasive even if it loses their tone (sometimes grammar/usage/typo edits can be unwise if they're not simply making understanding possible).

Which of course raises the question "Well who (aside from the original author) is to say what they intended? Isn't it a subjective call, when something crosses that line?" Yeah, it would be. I don't know that there's any better answer than "you know it when you see it."

Also helpful to the conversation: This 13-year-old Q/A from meta.stackexchange.com, from the formative days, pretty well sums it up:

You edit to make things better, clearer, more effective -- never to change meaning.

Answer 2

Your question goes to authorship and to editorship.
I think you have stepped over a line here, but only just, and for good reason!

Editing in substance that takes the answer beyond the answerer's intention is invasive editing — this is in the realm of community wiki (which the answer isn't intended to be).

It might have been a little better to leave the additional content in the comment; to have the reference in the related question directed to the comment. And, to also encourage the answerer to incorporate the substance into their answer themselves (thus rendering the comment redundant).

However, in this particular case the edit is trivial enough to be permissible/forgivable and should stand, but I do think it has crossed a line (just barely) between editorship and authorship.

Near duplicates are tricky
In some ideal world: the two question should be melded together seamlessly, but this would require both sets of questioners and answers (and commenters) to rewrite/tweak their contributions to fit together perfectly — this is somewhat unrealistic.

In another ideal world: some high rep / moderator type would take on the job themselves, performing a complete rewrite/re-scripting of a combined question and combined top answers. This would render both originals obsolete. And how useful would this really be?

Overall I think you have straddled the boundary of editorship and functionality very reasonably.

Answer 3

I am usually a stickler about "invasive" edits, which I take to mean edits which in some way affect the content of an answer. If an edit will in some way change the content of an answer, that edit probably shouldn't be made by anyone other than the OP.

But in this particular case the edit in question doesn't seem too worrisome to me. I think that there are at least a couple of reasons why this is the case. It does affect the content a bit, but really it seems to do this only by sharpening it.

Albeit, the sharpening is accomplished by adding some content. If the answer had been recently posted, I might feel differently. In that case it might seem better to leave a comment under the answer. The comment that provided the extra information was posted around the same time as the answer itself: over 8 years ago. That comment itself could have been converted to an answer and if I was contemplating this edit around the time that the comment and answer were posted, I would probably first urge the commenter to write an answer.

I also might feel differently if the added content were not already a part of the existing Q&A. That is, if you have some information that would improve an answer, you should just write an answer yourself instead of editing someone else's answer.

But from a historical distance this sort of edit seems like an effort to tighten the entire Q&A. I think that there should be a couple of caveats here. If you introduce some new content like this, you should be sure that the addition is correct. When you reference comments or other answers from the Q&A, you should include links; this helps readers find the source of the edits.

Here are some rules of thumb that I take from this:

1. edits shouldn't affect the overall content of answers

2. an edit that pushes this boundary should be made in service of the entire Q&A and should not introduce any information that is not already present in the Q&A

3. edits should not introduce any errors or confusion

4. edits that reference comments should include links to those comments