21
$\begingroup$

Recently, I have seen a lot of questions that essentially ask, "Is this proof correct?" or "Can you verify my work is correct?" like this one.

Often, especially when the asker's work is in fact correct, these questions have a one-word answer. I feel that these questions could be much improved by taking advantage of an underused feature: answering your own question.

Instead of asking the question and presenting your own work in the question, if the asker feels they may have a complete solution, they ought to post this as an answer instead of including the proof in the question (mentioning instead "I have included what I believe to be a solution, and would like to make sure it is correct"). This could be made clear in the FAQ or similar pages.


It seems that this question is related, and an answer there suggests answering your own question.

$\endgroup$
19
$\begingroup$

One problem with this is that often in these cases, the asker is not directly interested in finding a solution, but just wants reassurance that what they've done is correct, or to know where they've gone wrong. Otherwise they can always include a statement along the lines of "I would also be interested in other methods of solving it" at the end of the question. Asking and answering one's own question is likely to result in other users posting answers, possibly in a completely different direction to what the asker was thinking of, which is not what the asker wants.

Regardless, what I think is the main problem with this approach is that if the asker posts an incorrect proof, the explanation of the error, why it occurred, and how to fix it can be quite long. If answerers then want to address this issue, they would have to do so in the comments section of the posted answer (which is clearly not ideal) or by posting an "answer to an answer" which feels messy, to me at least.

Also, I think that the following thread is relevant: Asking questions with very short answers the community seems to agree (or did agree) that asking questions with short answers such as "is my proof correct" is perfectly valid, and doesn't need improvement. After all, there's nothing stopping someone answering such a question with "Yes, your proof is fine, but here's a faster/better/different way of doing it$\dots$" or "No, there is an error, you can avoid this by doing$\dots$".

$\endgroup$
  • 8
    $\begingroup$ In addition: A proof in an alternative answer to the self-answer might be so good that noone cares reading the likely cumbersome self-answer to check it for problems ... $\endgroup$ – Hagen von Eitzen Apr 6 '13 at 17:04
  • $\begingroup$ @HagenvonEitzen Exactly, I think that the focus would too often be drawn away from what the asker actually wants answering, and will instead become "How to solve the problem presented". $\endgroup$ – Tom Oldfield Apr 6 '13 at 22:02
11
$\begingroup$

The "is this correct" questions are actually useful. They show that the user has some reasoning process, and wants to be sure that he is not making mistakes, silly or otherwise. I believe that outlawing such questions would go against the purpose of this site, which is to help users answer their questions and to verify their work. Asking "is this correct" questions would help people spot reasoning errors.

$\endgroup$

You must log in to answer this question.

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