40
$\begingroup$

Today I went to ask a question, but I discovered that somebody else had already asked the exact same thing. Goody for me, right?

Well, the trouble is, the question got several answers, but I don't understand any of them.

If it were my question, I would probably add a comment or edit the question text. But this is somebody else's question, and it was asked a couple of years ago and hasn't been touched since. So I'm not sure what the best thing to do is. My options seem to be:

  1. Reanimate an ancient question with an "I don't understand this" comment and get yelled at for disturbing a long-dead question.

  2. Post a new question which will immediately be closed as an exact duplicate of the old one.

What do we think, folks?

$\endgroup$
  • 6
    $\begingroup$ In my opinion, it should be OK for you to edit that question asking for clarification. $\endgroup$ – user23211 May 3 '12 at 12:42
  • 22
    $\begingroup$ A comment will not bump an old question to the front page and probably won't be seen by anyone except the person whose question or answer receives the comment. Yelling at people for bumping old questions is reserved for those who bump by making trivial edits; substantive edits shouldn't upset anyone. $\endgroup$ – Gerry Myerson May 3 '12 at 13:17
24
$\begingroup$

It is fine to post a question of the form "Someone asked this here; but I don't understand this and that and that in the given answers".

Edit: As Arturo says, if the question has been idle for a while it's fine. Otherwise it might be better to leave a comment on one of the answers and ask for clarifications.

I also think that editing someone's question and asking for clarifications is rude, so I am always against that.

$\endgroup$
  • 1
    $\begingroup$ Then that is what I shall do... $\endgroup$ – MathematicalOrchid May 3 '12 at 13:30
  • 6
    $\begingroup$ It's probably fine if the question is pretty old and has not seen activity in a while. If it is recent and still active, I would discourage the practice. $\endgroup$ – Arturo Magidin May 3 '12 at 13:32
28
$\begingroup$

Given the current SE design, there is no optimal way to do this. Asking a new question only to clarify answer(s) of an old one creates an exact duplicate. Too many duplicates add great complexity to the site, adding extra work to classification (tagging, abstract dups), they complicate navigation of search results, etc. However, the alternative, bumping an old question, may not get your question full exposure, since many folks read only the newest vs. active questions.

Here is a proposed compromise. First, bump the old question by appending at the end a horizontal rule-delineated section saying why you bumped it, signed with your username, and date, e.g.


Bump: Why can one assume that the ring in XYZs proof is a UFD? --YourName, 2012/5/3


If possible, wait at least a few days for answers. If none arrive then, and only then, consider posting a new question. Hopefully such duplication should be necessary only in rare cases.

This presumes that the added complexity of duplicates is much worse to the site than is the thread bump - which seems a reasonable hypothesis, esp. if one views the site as an evolving body of mathematical knowledge, which we strive to organize. Thoughts?

$\endgroup$
  • 8
    $\begingroup$ Agree. More questions is more search results, which need to be human-parseable. More answers is more knowledge in a convenient place. $\endgroup$ – bright-star May 6 '13 at 6:41

You must log in to answer this question.

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