3
$\begingroup$

What to do when a question just wants a yes/no answer.

In the case of this question of mine, the answer was a 'yes' or a 'no - and here's why'. In the case of questions with the proof verification tag, I suspect this is standard. This means that if the proof is correct we get just a 'yes', which is the case in that linked question. Now I had read on this meta somewhere that the correct action to remove this from the 'unanswered' list is to merely throw a community answer onto it. This is what I had done, but it was deleted.

1) Should proof verification questions be formulated in some different way, i.e. "Is this proof correct, is there a better proof?". One problem I see with this is that it opens up duplicate tagging immediately, when the primary question is, "is my proof correct?".

b. Is it acceptable to post 'yes' as an answer, and should this be a community answer?

iii - Do proof verification questions not really fit the purpose of the site in the first place? It seems that they are a way to avoid duplicates and they do primarily cater to the user at hand. I can see them promoting interesting answers in the case they are wrong, but if a user is asking for a proof verification, they most likely do have a valid proof, or alternatively, they are either being lazy and not rigorously checking their own proof or they lack confidence in their proof, which they should be able to mitigate by thinking about it more and combing over it.

$\endgroup$
10
  • $\begingroup$ Advanced apologies for the tags if they are bad. $\endgroup$ Sep 27, 2015 at 10:11
  • $\begingroup$ Having trouble fitting the title and/or fixing the first line. $\endgroup$ Sep 27, 2015 at 10:16
  • $\begingroup$ If you have a look at related questions shown in sidebar on the right and if you browse other questions tagged (solution-verification), you will probably find out that many of the points asked in your post have been previously discussed on meta. $\endgroup$ Sep 27, 2015 at 10:31
  • 2
    $\begingroup$ About your CW answer which was deleted. I think if you prefaced it with something like: "I am posting Crostul's comment as a CW-answer, so that the question does not remain unanswered." it would make clearer to the reviewers what is going on. (And adding link to the comment in question and perhaps link to meta post with the advice about questions answered in comments would make this even better.) $\endgroup$ Sep 27, 2015 at 10:35
  • $\begingroup$ @MartinSleziak This seemed interesting to me: meta.math.stackexchange.com/a/5026/261481 Writing it as a regular question, and leaving the proof as a regular answer. But unfortunately 1) The question either will lack context or 2) It will be a duplicate. $\endgroup$ Sep 27, 2015 at 10:35
  • $\begingroup$ @MartinSleziak In regard to your second comment, I agree. Although I have never done the review process, so I don't know if it doesn't give you any option to see the post, but it seems lazy that they didn't even investigate at all. In regard to the edit, I still agree, thanks. $\endgroup$ Sep 27, 2015 at 10:36
  • $\begingroup$ I will add two more comments to what you wrote above. If you say directly in the question that you are posting your own proof as an answer and that you are mainly interested in verifying whether your proof is correct, then I do not think it will be closed for lack of context. Whether or not it would be closed as a duplicate is less clear. However, this answer suggests that such questions should be closed as duplicates only after the OP has received sufficient feedback on their solution. $\endgroup$ Sep 27, 2015 at 11:09
  • $\begingroup$ And as a last comment I will add that what is acceptable or not on the site evolves with time. If you have a look at point 8 in this post, mixedmath♦ says there that in their opinion question asking for verification of a solution will probably disallowed in the future. (And I can certainly imagine that too many questions of this type can become a problem for this site, so it is possible that we eventually get there.) See also Asaf's answer to the same question and discussion in the comments there. $\endgroup$ Sep 27, 2015 at 11:12
  • $\begingroup$ Sometimes I wish authors would simply put answers to every one of the problems for the benefit of self-learners because I worry sometimes if I broke $1000 + 1$ rules just by the act of posting on here :) $\endgroup$
    – user273143
    Sep 28, 2015 at 17:38
  • $\begingroup$ I want to note that noone appretiated(or noticed) my numbering: 1), b. , iii - $\endgroup$ Oct 3, 2015 at 6:49

0

You must log in to answer this question.