9

What can I do? Some users would say I need to accept answers.

Sometimes I don't know the topic enough to be certain which answer is the most correct. Also I don't want accepting an answer to mislead other users to think that's the most correct answer. (when sometimes an answer I accepted got 3 votes and a current or later answers get 12 votes and is in fact more correct).

Also from experience in the past, if 5 people answered the question, it may be natural that each person thinks he has a better answer. And some would message me why I accepted another person's answer but not his.

And then after going through lengthy process of explaining why, they might actually hate me and start downvoting my questions or future questions or act "whatever you do, I am against you" kind of attitude. What could I do?

  • 8
    Small note: the accepted answer is technically not "the most correct", but the one that helped you most. – Anton Menshov Jan 15 at 23:37
  • 1
    If you are worried that the answer you accept will stick it to the top forever, then be aware that there is a feature request that addresses such concerns at meta.stackexchange.com/questions/268666/… – PolyGeo Jan 15 at 23:40
  • 4
    Hate has no place in our Code of Conduct. If you see it, flag it. – PolyGeo Jan 15 at 23:43
  • 1
    "Sometimes I don't know the topic enough to be certain which answer is the most correct." If you are the one who asked the question, then you are probably able to test the solutions provided in the answers and see which ones work, right? – walen Jan 16 at 13:04
19

Flag any comments requesting you accept an answer or challenging your choice as "It's no longer needed. / This comment is outdated, conversational or not relevant to this post."

Such comments are noise and should not be posted in the first place. You have complete personal freedom to accept any answer you like for any reason, or to not accept any answer at all.

| improve this answer | |
3

From the accepting an answer FAQ, my emphasis:

Don't hesitate to accept an answer that is well-written, suggests a good practice and works for you.

If it works for you and was helpful, then it's an acceptable answer.

Even if it didn't work for you, accepting an answer is completely up to you. Upvotes from community members indicate when members felt an answer was particularly useful. Similarly, you can always upvote other answers that were helpful but did not fully answer your question.

Take this Meta question for example. The "answer" was a staff member's reply, but it was downvoted heavily. That answer was fitting in the eyes of the question asker, and therefore was accepted. You could argue the OP did that because they wanted to make sure it was seen, but it "answered" the question (discussion) all the same.

What I'm getting at is the same thing others have mentioned: It's up to you, and you don't need to explain yourself. It would be beneficial, however, if you chose the answer that worked best, or best suited the question.

| improve this answer | |
  • 2
    the statement "If it works for you and was helpful, then it's acceptable - end of discussion." sounded like after stating that, nobody else should have any opinion or any discussion. This sounded like some kind of ultimatum that I encounter at work or in daily life. If long time ago (or even now), one answer said, "it didn't work: use setTimeout(fn, 1000) to solve it" and it worked and was helpful, and is marked as the accepted answer, then if half of the people visiting the page in the future use this method instead of DOM ready, then it hurts software quality and ultimately, users – nonopolarity Jan 16 at 0:15
  • @nopole Just because it's acceptable doesn't necessarily mean it's the best answer, however. I'm in agreement with you that the best answer should be chosen, however that doesn't mean that in the instance you provided setTimeout(fn, 1000) is wholly wrong, right? I'm going to walk back my "end of discussion" part because you're definitely correct there, though. It's not the end of discussion. – Spevacus Jan 16 at 0:19
  • 1
    @nopole In such a case, you should probably edit the answer or add a bold **warning** in a comment. – maaartinus Jan 16 at 0:21
  • @Spevacus the use of setTimeout(fn, 1000) actually could be wrong, say, something didn't work, and a programmer found that one second later, or using setTimeout(fn, 1000), it "miraculously worked". So the engineer put that line in and considered it fixed. Then 2 months later, another programmer found that something didn't work, and setTimeout(fn, 1000) sometimes worked sometimes not, and so he put in setTimeout(fn, 1500) and now the page is full of patches and hacks, when the correct way really in this case is to listen on the DOM Ready to do things. tinyurl.com/y3qhqjec – nonopolarity Jan 16 at 0:27
  • 2
    @nopole Then that's where I would say you should take maaartinus' comment into consideration. If you found an issue with the answer then pointing it out is pretty important, especially if there's a use case where it's entirely invalidated. Heck, you might teach the author of the answer in the process and catch the problem before he has a coworker throwing him under the bus. – Spevacus Jan 16 at 0:29

You must log in to answer this question.

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