I often need to edit my answers several times to fix errors or maybe to answer more directly the questions asked.

But each time I edit my answer, the post will get a bit higher in the top questions page, as being changed recently. It feels a bit wrong to push a post with your own answer, just because you fix a sign or try to clarify. (On the other hand I don't have the time to check my posts often enough to feel really happy about it before posting them. There are always some flaws.)

Is it in impolite to edit too often? Will the post also get more attention again, if I edit an already accepted answer?

• You can find some related past discussions if you look at How much (self) editing is too much? and other posts linked there. – Martin Sleziak May 29 '18 at 4:55
• Edit it ten times within the first 15 minutes? No problem. Edit it once a week for 10 weeks? bad. Edit it whenever someone posts a comment asking for clarification, OK. Indeed, edit it whenever it is already back on the front page, OK. – GEdgar May 29 '18 at 13:18
• Props for asking! – Jyrki Lahtonen May 31 '18 at 18:41
• Perfect, with Martin's link to a previous discussion, all your answers and comments, I now have a good impression how to handle it and how also learned a bit more how the community works. Saying within these guidelines will be no problem. :) – Steffen Plunder Jun 1 '18 at 3:28
• I would say that if you are sincerely trying to improve your answer, and not just trying to bump your question to the front page (which of course you would never do), then go ahead and make the edit. If the answer can be improved, it should be. The art of writing is the art of rewriting. – littleO Jun 1 '18 at 20:02
• I appreciate your consideration for other users but on a media with so many users, each individual seeks only content of their own interest, and if indeed the content you have posted is of value to an individual, it really is expected that they ought to exercise the patience to allow you to edit and redevelop that content at your own discretion. – Adam L Jun 3 '18 at 16:30
• Looking at the quality of MSE as a whole, correct answers should beat annoyed heavy users (others won't recognize anyway) by far ... – tired Jun 5 '18 at 23:52
• I can't see any bad in it until every step is really a significant improvement. – peterh Jun 9 '18 at 22:39

Gerry Myerson's reply is factually correct, and I do not want to dispute it, only add to it: The goal of the site is to have a repository of clear, complete, and correct answers. Those answers will last a long time and readers will continue to extract value from them far into the future. If some people are annoyed in the process of developing the archive, that is unfortunate, but it is of much less importance, and it is also transient.

I like GEdgar's comment, which contains practical advice on how to achieve clarity and correctness while minimizing annoyance. But if the two conflict, clarity and correctness must take precedence.

Every time you edit an answer, accepted or not, you push it to the top of the front page (for those who view the site in such a way that Active questions come out on top). If you do this a lot, some of us will get annoyed and wonder why you couldn't have taken a little more care to get your answer right the first (or third, or seventeenth) time.

In my role as a moderator I encounter this problem once in a while. You see, edit number 20 triggers an automatic flag for us diamond bearers to handle. My response (spread out over a sufficient number of comments) is roughly the following:

• It used to be the case that edit number 10 automatically turned the question into a Community Wiki. Meaning, among other things, that you would no longer gain reputation points from future votes on the question. This rule has since been abolished, but frequent "bumping" still annoys many users, and you really don't want that kind of attention drawn to your post.
• We acknowledge that occasionally a post needs several rounds of fine-tuning (in particular when you work the kinks out of your LaTeX, language booboos, and other details of the presentation). To make this less obtrusive we created the sandbox. Whenever you foresee the need for several improvements use the sandbox to polish your post, and then copy/paste the final version here.
• If it is clear to me that the poster acted in good faith, may be they didn't know about the bumping effect, or it's otherwise clear that they were not seeking to hog front page territory, I will add a remark along the lines of No harm this time. Just something for your information. If it seems to me that the poster was, in fact, trying to get more attention to their post, then I will lower my voice by one octave.

It's best not to bump. Choose correctness over concerns about trivial refinement. If you're making a significant enhancement then bumping is not so bad. Having correct equations is far more critical than rewording an awkward but otherwise correct sentence.

Trivial bumps for attention is most unwelcome. Editing answers or comments within 5 minutes is fairly great, it still causes a bump but since it's not recorded in the 'Edit History' you're making less of a mess.

It's like the Murphy's Law of SE answers: "One must submit the answer to immediately realize it needs refinement.".

Compose using the answer box then cut and paste it into a text editing program where you can spell check, see 'the big picture', or simply save it while you think more or do some research.

I find that the tiny answer box is part of the problem, it seems to have a 'resize grabber' (bottom center) to expand the 'answer area' but that doesn't work on mobile. If you're on a MathJax enabled site that box is still quite useful and you can always see the preview below.

With a lengthy and complex answer it can be difficult to scroll up and down (on mobile) over images and links for fear of clicking on something - fortunately it asks if you're sure you want to leave the webpage.

I've had an answer box sit open for more than a day while I took time to research, eat, and rest; don't worry about it closing on you, in fact when my browser crashes (sometimes MathJax overload, combined with having over 100 tabs open) upon restart my answer is almost always repopulated into the box.

Some answers are difficult and involve a fair bit of research and confirmation, reasonably this can take more than one swing to get it correct. If can see that I have most of the answer(s), and the question is a week old and unanswered, I will post an unpolished answer with a disclaimer that most sites and people respect:

<sub> Note: I have proofread this but will be returning for another edit in a few hours.</sub>

Note: I have proofread this and it's finished, there's no need to misread what is written and make an unnecessary edit as has occurred already.

• To be honest I see the edit you mention more as a correction than vandalizing. (Since you are using MathJax for something which is perfectly achievable by Markdown and definitely is not mathematics. You could simply write <sub> inside backticks instead of using MathJax.) – Martin Sleziak Jun 4 '18 at 9:54
• Also I would be a bit careful about recommending using the feature of Stack Exchange that draft of answers are kept. I have read too many complaints of users who lost quite a lot of work to consider this reliable. (Although you have included in your answer that it is "almost always" kept - which perhaps counts as a recommendation against relying on this too much.) – Martin Sleziak Jun 4 '18 at 9:56
• The revision after my edit shows <sub>. Which seems to be what you wanted to achieve. And it also avoids unnecessary use of MathJax as in your original post - which is problematic for several reasons. (Note: This comment is a response to another comment by the OP - which was deleted in the meantime. Still from the - now deleted - comment I guess that clarification of my edit was needed.) – Martin Sleziak Jun 4 '18 at 10:13
• I apologise for a bad edit... but I thought you first had to approve of my edit before it came through... ? If you did not like it, why would you approve it? – Mr Pie Jun 5 '18 at 1:50
• @user477343 - I did not approve it, I came back a few hours later and it had been approved. I'm below the Rep to to have rollback/reject so the only thing I could do was edit it back. IF I had the reject ability I would have rejected the first edit (yours) for changing the meaning, the subsequent one can't be rejected for a trivial edit because the user has enough Rep to make trivial edits. Now an answer that was fine has three edits and over half a dozen comments (some deleted). It is a bunch of extra work in exchange for the UpVotes this answer has earned. Not angry, lesson learned. 🙂 – Rob Jun 5 '18 at 2:12
• @user477343 Only users below 2k have to make suggested edits (which then have to be approved through review) rather than directly editing posts - you are already beyond that. Moreover, on meta there are no suggested edits to posts; only on the main site. – Martin Sleziak Jun 5 '18 at 7:09

It's admirable that you try to be conscientious about this sort of thing. There are some people here on MSE I've had the displeasure of dealing with who use things like this to gain advantages in a very ugly competition. Sometimes they even use the rules to get other people punished.

If you edit your answer because you have definitely made a mistake that renders your answer thoroughly wrong, then you absolutely have to edit it. If you were trying to game the rules to get upvotes, that would be wrong.

My suggestion to you is this: when you're editing an answer that is already live on the site, take a couple of minutes to read it beginning to end, or at least skim it beginning to end. Maybe you'll catch instances of other things that need to be changed to account for the change that prompted you to edit in the first place.

Another suggestion: if you notice that another answer posts as you're typing the first version of yours, take that as a cue to slow down, not speed up. Though of course there is the danger others will accuse you of plagiarism, they don't want to believe that you could have come up with the same thing at the same time as someone else.

These are just suggestions.

If you start feeling that you are not going to be happy with your answer before a number of edits, and you are concerned about annoying active-questions aficionados, what you can do is delete temporarily your answer. You can still edit a deleted answer of yours as many times as you like, it won't show up anywhere. This might have side effects that I am not aware of, but it is something that I sometimes do, and I haven't noticed anything suspect.

Edit: apparently, editing a deleted answer still bumps the question. Nevermind then. I will still leave this answer in case anyone get the same idea. And I swear (I shouldn't, I know) this is the last time I edit this answer. :-)

• Did you mean to say temporarily delete your answer rather than your question? Notice that even if the answer is deleted, editing the answers still bumps the question. – Martin Sleziak Jun 3 '18 at 0:43
• @MartinSleziak Yes I did mean that, thanks. Oh really? That makes little sense, it would probably make sense to change that. – Arnaud Mortier Jun 3 '18 at 0:45
• Well as you can see in the link I've included in my previous comment, there is a feature request about this and it is marked (status-bydesign): Stop an edited deleted answer from bumping the question. I will add that the FAQ lists among reasons for bumping: "Editing the question or any of its answers (even if the answer is deleted)". Here is link to the relevant FAQ item: What can cause a question to be bumped? – Martin Sleziak Jun 3 '18 at 0:49