I recently was looking up a question and noticed a small improvement I could make to an answer by adding an additional detail. The user whose answer it was told me that I should not have made this edit because the answer was a few years old.

Was this user correct? Are older questions and answers on the site not supposed to be edited? And if so, why not?

Edit: The nature of the edit was citing a theorem to explain an assertion. The assertion would be obvious to someone familiar with the material, but a novice would probably find the citation helpful. Hence, the edit was not merely a copy edit, but it did not substantially change the content of the answer, only the presentation.

  • $\begingroup$ You should be accurate. It is because the post is very old and the fact that the edit is extremely minor. $\endgroup$
    – Asaf Karagila Mod
    Commented Mar 3, 2014 at 3:34
  • 1
    $\begingroup$ For what it's worth, I voted to reject the edit. I don't think it's necessary, and edits to very old posts should be substantial improvements or fixes to something actually broken. In general, I oppose edits of content for such old posts. $\endgroup$
    – user61527
    Commented Mar 3, 2014 at 4:22
  • 1
    $\begingroup$ @T.Bongers I thought it would be helpful because when I first read the answer, I did not understand the reason for the assertion. Consequently, I looked it up and improved the answer, to help future readers in my position. $\endgroup$
    – augurar
    Commented Mar 3, 2014 at 4:34
  • 2
    $\begingroup$ I haven't looked at the question/answer --- perhaps you could have achieved the same goal by putting your clarification in a comment? $\endgroup$ Commented Mar 3, 2014 at 6:33
  • $\begingroup$ Here is link to the suggested edit mentioned in the post. $\endgroup$ Commented Mar 3, 2014 at 7:44
  • 7
    $\begingroup$ Perhaps a better course of action here would have been to add a comment below saying what you didn't understand and that Cantor's Theorem resolved it. Saves the bump, and anyone reading the post will read the comments (it would have been the first comment). $\endgroup$
    – user1729
    Commented Mar 3, 2014 at 10:43
  • 1
    $\begingroup$ Alas, there is little collaboration on MSE. If the edit is minor, it is usually rejected as being too minor to warrant a bump. If it is major, it is usually rejected as speaking too much for the author. So we rarely get to benefit from the collaborative tools offered by the SE platform. Note that the author receives notifications of every edit, so the author can revise the edit as need be. I think we should try to be more open minded about such capabilities. Imagine, for example, the beauty of a collaboratiion by Georges Elencwajg, Matt E, and Pierre-Yves Gaillard. $\endgroup$ Commented Mar 3, 2014 at 15:45
  • $\begingroup$ @Bill: I think that this sort of collaboration works best when all sides agree to collaborate; not when each side adds some information and the other edits it. $\endgroup$
    – Asaf Karagila Mod
    Commented Mar 4, 2014 at 6:05

1 Answer 1


If you can improve a post, go ahead and do it. Just stay clear of the posts by that particular user.

Whether the post is 3 weeks old or 3 years old makes no difference in practical terms; either way, the question was already off the front page and will be bumped there. Which is fine; this is additional exposure and peer-review for the content. While many consecutive edits to old posts can disturb the delicate harmony of the front page, this isn't a problem if you edit at most a couple of posts per hour.

  • 2
    $\begingroup$ I can improve many posts by many users. This site is sort of a trace of someone's knowledge online. Especially when the user is using their real name, I find content modifying edits to be a bit much. Both minor and major. $\endgroup$
    – Asaf Karagila Mod
    Commented Mar 3, 2014 at 3:39
  • 5
    $\begingroup$ (The point in my previous comment is that there should be a difference between actual content editing, and style and grammar editing, and formatting editing.) $\endgroup$
    – Asaf Karagila Mod
    Commented Mar 3, 2014 at 4:13
  • 1
    $\begingroup$ A couple per hour? I prefer the limit of 3 per day for edits of stale questions. $\endgroup$
    – GEdgar
    Commented Mar 3, 2014 at 15:07
  • 2
    $\begingroup$ @GEdgar "3 per day" sounds like a MathOverflow kind of number. Currently, Math.SE gets 530 new questions every day, about 800 new answers (extrapolated from data here), and this is before we get to edits of questions and answers by their authors (which are the most common kind of edits)... Also, while a 2011 question may be stale for old-time users, note that 85% of current users joined the site in 2012 or later. For them this is not-seen-before content when it bumps. $\endgroup$
    – user127096
    Commented Mar 3, 2014 at 19:20
  • $\begingroup$ @GEdgar Here are some numbers (although the site has grown since then): How much bumping is too much? $\endgroup$ Commented Mar 4, 2014 at 15:17

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .