How much (self) editing is too much?

Question

(N.B. not to be confused with this question)

We are all aware of editing being a core value of the part-Wiki nature of the Stack Exchange websites.

We are also all aware of the website providing a system of check-and-balances for edits, as a way to prevent and counteract intentional vandalism or spam content. One of the most prominent of this aspect is that every time and question or an answer is edited, and every time a question is re-tagged, the software 'bumps' the question up to the top of the list of "Active" questions.

To prevent a user from using the above safety check as a method of keeping a question constantly on the front page (and presumably earn more reputation from the increased exposure), a secondary rule is in place that a post when edited too-much (by too many editors or by the original poster 10 or more times) becomes automatically Community Wiki. One can think of this as a way of reducing the likelihood of gaming the system.

Now, given the above check-and-balances (which I think takes pretty good care of the technical and reputation aspect of the problem), is there any other reason why we should discourage "frequent and plenty" (see below the cut) edits?

One possible objection I can see that may be raised is that "repeated edits crowd the front page and reduces the visibility of other questions". Some may counter that the existence of the "Newest" page in the list of questions should amend that. But for better or for worse recall that the default front-page to a first time user is the "Active" page.

---

For the sake of argument, let us define "frequent and plenty" thus: imagine a user who decides, for whatever reason, to write his or her response piecemeal, adding to it with a new edit every half hour, over the period of say 8 hours. (So 16 edits to the post in one working day.)

Further clarification: this question is not about the occasional bug fix; nor is it about discouraging edits in the abstract. It is only about a large number of edits coming in a short period of time. To phrase it differently: given that the software already "penalizes" (for the lack of a better word) a user for doing too many edits, is it then ok for a user to knowingly make "frequent and plenty" (see above) edits provided he/she is willing to "pay the tax" (so to speak)?

Answer 1

(Note: People in chat have pointed out to me that this meta question might have been an attempt to continue a comment discussion from this mainsite post -- at least Willie pointed the particpants in that question to this one, and it certainly appears that those participants want this meta question to be about whether that particular user's editing patterns are acceptable or not. However, my answer here is an answer to the general question Willie actually asked: Is there a reason to discourage users in general from making many edits to a post over a short time, on the grounds solely that the number of edits is too large, and independently of their content?)

---

If the "frequent and plenty" edits actually make the post better, then the only problem I can see they create is the frontpage bumping. This problem should not be solved by discouraging the improving of posts; see below. On the other hand, edits that make the post worse should be discouraged no matter if there are only one of them. For both kinds of edits, their absolute number is irrelevant.

I fairly frequently amend my answers as I find simpler way to explain or define things or am reminded of considerations I should have taken into account in the first place. I also routinely edit my answers to fix typos and bad prose, because I want my answer to be the best possible answer I can provide.

I don't recall that I've ever hit the CW limit with this, but I still think it would be bad to actively discourage edits. The alternative would be to advise users either to leave improvable answers in a non-improved state, or to sit on an unsubmitted answer until they were 100% certain that they couldn't possibly ever find a way to improve it. Neither of these would be beneficial to our primary mission of providing high-quality answer in a timely and helpful manner. (The latter outcome would be harmful through duplication of effort; it would cause a much larger risk of writing the same answer that someone else has already completed and is just setting aside on the off chance that he would find a way to improve it before finally submitting).

I can see that a stream of rep-farming inconsequential edits can be annoying, but I think a better fix for that would be to tweak the bumping algorithm. For example, there could be a rule in the software that edits from a single user cannot cause a post to be bumped more than once every 10 hours (for some appropriate value of 10). Then if one edited the post quickly after having bumped it once, the bumping would simply be deferred until 10 hours after the last bump.

That way, monopolizing the front page would be harder, but self-vandalism edits would still be caught eventually. Classic drive-by vandalism of a post that haven't been edited for a while would be caught immediately, just like with the current behavior, because the bump would only be deferred if there had been a bump recently.

Answer 2

In my view, repeated editing to correct typos, or mathematical mistakes are fine. But at some point the answer should be about done, usually about 5-6 edits, and in some extreme cases 10 edits.

Rarely there is an answer which is edited many times and each edit adds on new information, bibliography and citation (I recall an answer or two of this nature on questions of historical nature).

However writing an answer and editing it 30 or more times simply to expand the proof and add something which results as a piece of lecture notes, or a blog post, this seems unreasonable usage of the software in my eyes. If you wish to work a complete and detailed answer, do it in advance rather than during the post itself. I can personally testify working hours and days on some of my answers before posting them.

One other alternative is to write a blog, or actual notes, on the topic and link them in an answer. However simply linking to a blog or notes is itself an unreasonable answer, instead a summary of the argument and a direction towards the complete and detailed topic should be written.