13
$\begingroup$

If I have a question that depends on a previous question that's already resolved, how do I post that follow up Question?

  1. Edit my original question
  2. Post a new one
| |
$\endgroup$
  • 5
    $\begingroup$ If it's already resolved, radical changes to the original question would waste all of the effort of the people who wrote back to you. So don't do that. In other words, "Do not move the goalposts when it is cheap and easy to make new goalposts." Then again, do not run wild with too may goalposts: make sure your questions are carefully written :) $\endgroup$ – rschwieb Aug 12 '14 at 14:06
  • 3
    $\begingroup$ Small suggestion: a single question mark suffices as an indication that a sentence is a question. $\endgroup$ – user147263 Aug 12 '14 at 14:34
  • $\begingroup$ @900sit-upsaday Ya I will try not to do that again. $\endgroup$ – Sudhanshu Aug 12 '14 at 14:39
  • 1
    $\begingroup$ Very similar question just came up at MO: meta.mathoverflow.net/questions/1845/… $\endgroup$ – Gerry Myerson Aug 12 '14 at 23:42
19
$\begingroup$

Post a new question, linking back to the original with a brief explanation of how they are connected.

Editing the old question is bad form, as it would require the existing answers to be re-written (which is a pain for those kind people who took the time to answer the question!).


There are, of course, exceptions. The main one, as has been pointed out in the comments, is when "there is a trivial case you forgot to rule out (such as the function is identically zero) and the answer is merely that trivial case. Then, feel free to orphan the trite answer by improving your question."

The key points here, thought, are that the case should be trivial and the answer should be trite. For example, The OP of this question omitted a non-trivial case, and the answer was simple but pleasing (it was not trite$^{\dagger}$). So the OP started a new question rather than editing the original.

$^{\dagger}$ Trite (adj.): (of a remark or idea) lacking originality or freshness; dull on account of overuse.

| |
$\endgroup$
  • 5
    $\begingroup$ I agree. The exception is when there is a trivial case you forgot to rule out (such as "the function is identically zero") and the answer is merely that trivial case. Then, feel free to orphan the trite answer by improving your question. $\endgroup$ – GEdgar Aug 12 '14 at 16:29
  • $\begingroup$ Is it possible to lock/close a question where the question is now outdated / obsolete as a result of the follow-up? $\endgroup$ – dustytrash Dec 5 '19 at 22:08
  • $\begingroup$ @dustytrash well, there is a benefit to keeping such questions. You can flag the question if you want. Is there a specific question you are thinking about? $\endgroup$ – user1729 Dec 6 '19 at 8:41
  • $\begingroup$ Also, it's nice to keep old questions which have an answer. Answers can take a long time to figure out, and then a long time to compose. Deleting them because they have been superceded by something else feels dismissive of this effort. $\endgroup$ – user1729 Dec 6 '19 at 8:44

You must log in to answer this question.

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