This is meant as a conclusion (and possible continuation) of a comment-discussion I had with @Daniel around this answer. I have a strong opinion on this myself, but I'm here for a very short time and so my intuition may well be misguided.

Now my question:

Assume a question has an accepted answer for a long time. Then (say more than a year later) someone posts a new answer (maybe even because the package under consideration has changed since the time of asking) which the OP thinks to be better. Should he accept the new, better answer even though the person posting the new answer has much more rep than the person who posted the answer which used to be accepted before?

I have seen this question but the results seemed inconclusive, so I'd like to see some more opinions.

  • 5
    Note that @user inside an answer or question doesn't trigger a notification and should be avoided. It only works within comments and the chat. For comments the user must have posted a comment on that very same post before in order for this to work. Commented Mar 29, 2012 at 8:48

1 Answer 1


Officially the OP should accept the answer which was the most useful to him/her. This doesn't always need to be the most technically correct answer. However, this can be helpful for other people who have the same issue. There can be multiple "correct" answers, even if they are different. They should of course all be up-voted. It is OK for the asker to change the acceptance to a more recent answer.

The amount of reputation of the answerers should not be used to decide that. People with low rep on this site might be actually (La)TeX gurus which haven't been around here for long or just not very often. Also, high rep users shouldn't get discriminated against just because they already have more rep.

Changing the accepted answer can be a little frustrating for the author of the former accepted answer, but shouldn't be a big deal. It can happen that the original answer was a complicated workaround to a then current issue and the answerer spend quite a bit of time on it, only to be "replaced" by a "use the new version X.Y of package Z"-type answer a few days, weeks or months later. However, this is just life and always the risk. You should still have a decent amount of up-votes on the answer.

What I don't like is un-accepting a very helpful answer only to self-accept the conclusion answer of the OP, based on the other answer. This happened to me once, for example.

  • 6
    Yeah, you're last point is super-annoying. No dignity.
    – morbusg
    Commented Mar 29, 2012 at 21:23
  • Thank you very much for the clarification. Commented Mar 30, 2012 at 8:06

You must log in to answer this question.

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