I posted a question here and got two equally good answers. Both of them are short, useful and to the point. One person's answer is more upvoted than that of the other which I believe was due to the former answering much earlier than the latter. Both of them responded equally to my comments and doubts regarding both methods, thus putting more effort into it. Now I am in the moral dilemma of whom's answer I should accept?

  • If it's a tie, I think not giving the checkmark to one rather than the other is acceptable. Simply explain your dilemma to both answerers and thank them for helping you.
    – jub0bs
    Commented Jun 3, 2013 at 9:13
  • 4
    I guess you didn't use both answers for your initial purpose, why not thank the one who actually helped you? Additionally, as stated by Jubobs, you can thank explicitely the answerers. I personally see accepting more as a semantic flag "this question has a definite answer", than as a reputation reward or a "thank you" for the answer itself. The "thank you" is given by your upvote and maybe a comment, while the rep. reward is given by the upvotes (sometimes complemented by a bounty).
    – T. Verron
    Commented Jun 3, 2013 at 12:59
  • 1
    Also, I seriously doubt any of the answerers would question your choice. They know that both answers are equally good, and they know that you can only accept one of them.
    – T. Verron
    Commented Jun 3, 2013 at 13:04
  • 2
  • 1
    No worries, we just don't care, in fact it's the common theme of teasing each other in the chat room. You can join us too if you like.
    – percusse
    Commented Jun 3, 2013 at 19:07
  • I think It's fully in your control and you can be better judge. Leaving appreciating comments to other answers like Peter suggestion is good idea. Commented Jun 4, 2013 at 3:39

2 Answers 2


The answerers know that you can only accept one answer; they shouldn’t mind when you do so.

On english.stackexchange, I’ve been an answerer in this scenario many times, and I’ve never seen anyone unhappy when one answer among two or several good ones was accepted. If you like, a comment like “Thankyou, this [was very helpful/works fine/is a useful alternative], and I wish I could accept it as well as X’s answer” never goes amiss — while “thankyou” comments are deprecated in general, in this situation it gives useful extra information for readers.

  • I believe your answer is on the presumption that I am confused whether one answerer will be offended if I choose the other. But the problem is not that. The question is on what basis should I choose one of them. Since comparison of helpfulness and technical value is ruled out, what should be my next point of comparison? Should I just toss a coin then :-) ? Commented Jun 4, 2013 at 13:37
  • 1
    What I mean is more: if your choice is a bit arbitrary, that doesn’t really matter — it isn’t a big moral dilemma. So yes, you can flip a coin if you want; although I’ve never seen two answers so exactly equally good that I didn’t have some slight leaning towards one of them. Commented Jun 4, 2013 at 15:26
  • you are right!, I did have some inclination towards one of the answer as I thought it was more cool and which I accepted as an answer. But, then still the question remained for a hypothetical situation other than that of mine, where they are exactly equally good. Thanks for the reply. Commented Jun 4, 2013 at 15:47

If the two answers are absolutely of the same value, I think it could be a good practice to accept the one of the user with the lowest reputation.

In general, you should vote/accept the answer, not the answerer, but when there are two equally good answers, I think it is better to reward the user with the lowest reputation. It could be a way to make him/her more affectionate to TeX.SE!

  • Better is to not display reputation at all. Reputation display decreases the quality of the site, because it infantilizes its users. In the same vein, there should be a delay of 24 hours before the submitted answers actually show. 1) this would increase the long-term quality of questions, avoiding the build-up of "oh gosh I don't know what to do, I have never ever read any docs, let's ask on tex.stackexchange", 2) this would allow the OP to compare more serenely the answers and not validate the first one which happens to solve his/her problem.
    – user4686
    Commented Nov 12, 2017 at 16:01
  • @jfbu I don't look at reputation with such a pessimistic view, it could be also a (raw, of course) method to measure the answerers' reliability. 1) I don't either agree with the idea of the delay in submitting answers, our first goal is to help users, as quickly as possible. 2) On the other, I agree with the fact that users often accept the first answer, not the best (and more often don't accept answers at all) but the delay is still not a good idea, in my opinion.
    – CarLaTeX
    Commented Nov 12, 2017 at 16:16
  • in most cases, the user will accept the first answer which "solves" the problem; but the user was not qualified to solve the problem him/herself, and sometimes, its estimation of what actually has been solved might be erroneous. The current functioning of the site is not one of educating the public. It does not help increasing the average understanding of LaTeX and friends by users. I am quite convinced an objective assessment of quality of questions will conclude to a decline over the years. A delay of 24 hours would help reduce the user base to a less consumeristic one.
    – user4686
    Commented Nov 12, 2017 at 16:33
  • @jfbu A delay of 24 hours would help make the users ask elsewhere :)
    – CarLaTeX
    Commented Nov 12, 2017 at 16:35
  • precisely my point : ). There is a small community here of experts, including probably 3 or 4 (make that 10 ;-) ) among the top few world connoisseurs of LaTeX, and I find it a bit wasteful to observe them reply to questions which nowadays, very often, are from people who did not make any effort to read even a modest amount of documentation.
    – user4686
    Commented Nov 12, 2017 at 16:42
  • @jfbu That is really a pity!
    – CarLaTeX
    Commented Nov 12, 2017 at 19:58

You must log in to answer this question.

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