How to handle suggested edits that are correct but extend the answer or question drastically?

Question

There's just been an approval of an suggested edit which adds some true statements to the answer, but actually introduces some drastic changes to it.
The edit was introduced by an anonymous user. The original poster did not approve the edit.

There's the option to reject the edit for this reason:

This edit deviates from the original intent of the post. Even edits that must make drastic changes should strive to preserve the goals of the post's owner.

And indeed someone did choose this option.

However, there were also some approvals and eventually the edit was carried out.

Obviously, the edit doesn't add any harm. Quite the contrary, it improves the answer. And though, I don't think the edit is acceptable. It's not necessarily the opinion of the original poster.
In respect to this very edit, it's interesting but not important to say that the terms mentioned in this post have already existed at the time of the Deutsche Mark. And completely superfluous is the mentioning of Groschen, not least because it's an outdated term.

How should we handle "good" edits that add content to the post? Should be approve or reject them? Or should we skip them and wait for the original poster?

If such an edit got approved but wasn't approved by the original poster, should we rollback the edit?

Answer 1

In the concrete example, the edit adds only useless noise. It's not wrong, but unneccessary and should be abandoned.

In contrast, deleting such a paragraph would be too rude. Such changes can be suggested in comments, but if they don't correct a mistake or fill in necessary information, they shouldn't be made.

Even a lousy street slang should be left in its shape, if correct and acceptable, because with the name and logo, the person who rose the question is closely associated with it and should be threatened as the owner of it - except it is against the rules.

Answer 2

First of all, some statement on my rejection of the edit that sparked this discussion:

Regarding the usage of Fuffi, the author of the answer claimed that he mostly heard the term on TV. The edit makes a different claim, by saying that it is “colloquial in parts” of the German area. Due to their vagueness, these statements are not exactly contradictory, but they are not really compatible either. Thus, the edit is in the grey area between extending the answer and changing a (non-sidenote) statement – with the latter being something that would make an edit be a clear case for rejection (see below).

In addition to that, that edit added some stuff that does not really answer the question and also is a rather strong extension of the answer. Also note that the edit could have very well been a standalone answer. Taken all these aspects together, I considered it better to reject the edit. However, I do acknowledge that this edit was not a clear-cut case.

As to handling such edits in general:

I do not think, that skipping a review to wait for the OP to decide would actually work. Any suggested edit alerts all users that can review it in the top bar (which is good, since those should be handled quickly in most cases) and thus somebody having an opinion is bound to come along soon. If the OP is online at the time of the edit, he will most probably see it anyway – unless two people are very quick at rejecting that edit.

However, when rejecting an edit that we want the OP to know about and we have reason to believe that the OP does not see it if it gets rejected, we can consider leaving the OP a comment with a link to the suggested edit or whatever aspects we consider worth their consideration.

In general, some guidelines I would suggest on edits that add correct information to answers (with no claim to exhaustively cover other cases, and assuming that no other guidelines are violated), sorted by preference:

• If the edit changes any remotely essential statement made in the answer (i.e., pretty much anything that is not a sidenote), reject it.
• If the edit adds essential statements or arguments, reject it. (In this case, the edit could most probably be a standalone answer, see below.)
• If the edit extends an existing open list of examples¹, exceptions or similar, approve it. Also consider approving, if the list is claimed to be complete, but the added element is of minor relevance in comparison to the existing ones (e.g., an exception that only applies to liturgical texts nowadays).
• If the new (or corrected) content could be made into a valid standalone answer, reject it.
• If the edit adds examples¹, illustrations or similar that help understanding the answer, approve it. (Be careful with answers though whose main purpose was to explain something.)
• If the edit adds sidenotes¹, approve it.

---

¹ and includes no changes that are more radical.