20
$\begingroup$

Background:
I have, in my own biased opinion, an amazing and beautiful answer that was given to a question that was changed so that the answer was eventually made irrelevant. For the original question, in my personal opinion, it was an excellent answer.

The question:
Is it inappropriate to make a new question, based on an earlier version of the original question, and supply my answer as an answer?

More thoughts:
I really think the answer was a thing of beauty and I put some good time and thought into it. I have received quite a few up-votes on much simpler answers to the same general topic and I feel that this answer would be valuable to the same audience. I think that given the nature of the current question, that the answer will never "see light of day" or be found and accessible.

UPDATE: (request for clarification)
If I am going to ask the question, then answer it, how long is a reasonable time to wait for others to answer? I am committed that if someone else has a reasonable answer to select theirs as best before putting my own answer "into the ring".

$\endgroup$
  • 11
    $\begingroup$ You can post your answer simultaneously with your question. There is a feature to let you do that. I see no reason why you should wait before adding your answer. The opportunity to post your answer is the reason for asking the question in the first place. (You don't have to accept your answer, though.) $\endgroup$ – gung - Reinstate Monica Jan 8 '16 at 2:15
27
$\begingroup$

If you have a good answer to a question that's no longer there, it would be a great pity to leave things as they are, rather than have the benefit of your effort available to be easily found.

This is a problem that does come up a fair bit; generally once a good answer to a decent question that was posted appears, the poster should normally do no more than "clean up" the question (small edits); unless the answerer completely misunderstood the original question, the answered question should left asking whatever it was asking before.

There's a number of possibilities here:

  1. If nobody has yet answered the current version of the question, it can be rolled back (you can do it or you could flag it I guess). The OP can then ask a new question if they want to ask something different.

  2. If that's not suitable for some reason, you should feel absolutely free to post the old version of the question from the edit history as a new question (you must, however, credit its author and you should link to the original post).

  3. You could write an entirely new question to which your previous answer is an answer; in some cases this may be a better choice.

$\endgroup$
  • 1
    $\begingroup$ Thanks. Much improved. I have in turn edited that into the numbered list above it, which I probably should have done before. $\endgroup$ – Glen_b -Reinstate Monica Nov 26 '18 at 22:55
8
$\begingroup$

I want to agree with Glen_b here, but more emphatically. Once a question has been asked and has received an answer, it is not appropriate for the OP to change the question. That wastes the time of the person who answered, and gives the false impression that they did not answer the question correctly. Once answers have come in, edits to the question should be limited to refining it to be clear; it is not appropriate to change the substance of the question in a way that renders existing answers defunct. If the OP wishes to change the substance of the question then he/she should instead ask a new question.

I think the fairest thing here is to flag the inappropriate edits to the question and have moderators roll-back the edits so that the original question stands. This could be augmented by a comment to the OP recommending that they ask a new question. Although Glen gives some other possibilities in his answer, as second-best cases, I am of the view that the edits should be rolled-back.

$\endgroup$
  • 1
    $\begingroup$ Sometimes it is a good idea for edits to make evident both the original and the revised question; similarly the original and revised answer. But this is all case law. The over-arching principle must be what would be interesting and useful to others? $\endgroup$ – Nick Cox Nov 26 '18 at 9:32
  • $\begingroup$ Even in the case you're referring to, it would be just as easy for the OP to ask a new question rather than changing the substance of their existing question. That would be preferable in my view. Creating new questions is essentially costless, and it has the advantage of leaving both questions; the original badly-worded question remains, along with an answer that points out the problems, and then a new question is asked with greater clarity, which can then receive a substantive answer. $\endgroup$ – Reinstate Monica Nov 26 '18 at 9:33
  • 2
    $\begingroup$ I've experienced an answer of mine being made redundant if not ridiculous in roughly this way. (1) OP asks a very confused and/or incomplete question perhaps with errors. (2) My answer explains what's wrong (and often what else is needed). (3) OP sees the error of their ways and revises the question, so I have to rewrite my answer. Although it's temporarily irritating if not annoying to have one's answer made irrelevant, such a process has to be seen as positive long-term. So I don't agree with a blanket declaration that "it is not appropriate for the OP to change the question". $\endgroup$ – Nick Cox Nov 26 '18 at 9:35
  • 1
    $\begingroup$ Ben's comment refers to the comment that appears below his. (I corrected a mistake in wording.) $\endgroup$ – Nick Cox Nov 26 '18 at 9:36
  • 2
    $\begingroup$ "just as easy for the OP to ask a new question rather than changing the substance of their existing question". Perhaps, but your suggestion again seems dogmatic, that there is only one right way to proceed. Long term, it would often be messy to have two versions of the same question from the OP. As a long-term member I know some fraction of my time here will be wasted; naturally I want that fraction to be small but I won't ever get the fraction down to zero. People really are allowed, indeed expected, to improve their questions whenever possible. $\endgroup$ – Nick Cox Nov 26 '18 at 9:41

You must log in to answer this question.

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