When is an "edit" a separate response?

Question

With reference to this response that was edited recently:

The editor felt that the response (marked as accepted long ago by now) needed editing in terms of content (not presentation) but from my point of view, the "edit" should have been left as a separate response.

The edit has basically altered two characteristics of the original response:

1. Style of writing:

   • "But don't compute variance of a plain image, though! Better compute the variance of laplacian kernels"
   • "( maybe with some stride to save time)."

   I do not write like this.

2. Clarity / Accuracy of Content

   • Improvement by including a citation in the first paragraph which is inline with the guidelines.
   • Unclear in terms of contribution on the point of variance of Laplacians to infer focus.
     1. A Laplacian can be seen as a high-pass filter. Obtaining the variance of the Laplacian provides an estimate of the 'smoothness' of high frequencies. The edited response now concludes with Of course, another way you could be judging the blurriness of an image is the ratio of the power of the high frequencies relative to the lower frequencies. In this case, instead of obtaining the standard deviation of each MxN window, you would have to obtain the 2D FFT. This is slightly more complicated though..
     2. The point about "slide the window in strides for quicker processing" omits the impact this will have on the already high variance of the Laplacian.

This is not how I would choose to talk about this subject.

I think that the edits actually constitute a different response which could have been left as a separate answer, however the edits were accepted and the text, although attributed to me too, is not a response I feel 100% inline with.

What is the general consensus about this? Should the "severe editing" limits be spelled out explicitly too? As a user, I was notified about the edit but I was only given one option, "Approve", which did not seem to be making a big difference anyway...(Whether I "Approve" or not, the edits were appearing), have I missed something here? Would "flagging" the edit be an appropriate response?

This question is about clarity. I don't want to edit the response back to what it was myself or start an argument with anyone but at the same time I can't just let this slip.

Answer 1

I think the edits would probably add more value as a separate answer. They attempt to introduce (and remove) a little too much information, as opposed to just clarifying the existing answer. Unless one of the other mods (or the community) objects, I think it's fine if you just revert the post, and leave a comment for the editor to elaborate their thoughts as an answer instead.

As for how to handle these situations in general, I'm not sure if there's an official protocol, but I'd be fine with people flagging approved edits for moderator review when they're in a grey area (i.e. "in need of moderator intervention"). Sometimes edits look reasonable and get approved, but if something is controversial to the OP, a second look should be warranted.

As for the SE interface for edit notifications, I'm not sure if there's anything we can do about that. Most of the time edits are relatively mild, so flagging the exceptional cases should be sufficient.