Low number/percentage of accepted answers

Question

I am under the impression that too many questions don't have accepted answers, even though a lot of them do have answers (even more than one). I see this practice negatively for two reasons:

• Why answer questions if users don't bother to ackownledge that they read and selected an answer?
• It gives the impression that this website is just a free translation service where people can come and write a word or phrase, have it translated and then forget about the whole philosophy of the StackExchange community (giving feedback, commenting, accepting answers, etc.).

For example, here is a decent question (by the number of upvotes) with a decent answer (again, by the number of upvotes), but the answer has not been accepted so far (at the time of posting this). Is this because the person who asked the question does not consider it correct, or "good enough"? If that was the case, then that user should comment, asking for more details, alternatives, etc.

Obviously, that's not the only question with this situation. I just wanted to use it as an example.

What is your take on this? I think that having more accepted questions would show that the website is actually serving its purpose, which might be encouraging for new users trying to see a difference between this and a dictionary.

Answer 1

Accepting answers is definitely a major part of the StackExchange community. The FAQ entry says:

Which answer should I choose?

• Don't hesitate to accept an answer that is well-written, suggests a good practice and works for you.
• Otherwise, even if there are answers that are good enough but that you're not entirely satisfied by, you might wait 24 to 48 hours to give other people a chance to give you a better answer. A question with an accepted answer isn't as likely to receive further attention as one without an accepted answer.
• Make sure that besides working for you, the answer is really good practice. Sometimes after the answer gets accepted, another comes in, uncovering the fact that previous one was in fact a bad hack.
• The bottom line is to accept the answer that you found to be the most helpful to you, personally.

What should I do if none of the answers is suitable?

• Re-read your question critically and make certain that it's clear what the problem is, and what the specific question is. You might want to place a bold single sentence version of your question at the bottom of your post to make sure it's clear what's being asked.
• Re-read the answers. Consider accepting the best answer of the bunch, even if none of them are exactly what's needed.
• Wait a few days after asking it, and then edit your question with what you've learned since asking it — perhaps additional information will help others help you. This will also have the effect of bringing your question back onto the home page where fresh eyes will see it.
• If you have more than 75 reputation, you can add a bounty two days after the question is asked if you haven't accepted any answers yet. This will add incentive for people to answer your question.
• Eventually you'll resolve the problem or find a work-around (which may be, "This won't work, so here's what I did instead"). Please add an answer with this information and accept your own answer so that others with your same question can see the path you eventually followed.

Personally, this is how I've been approaching accepting answers on this site:

• After asking a question, I generally won't accept an answer for a week or two (unless a really excellent answer comes along) to give everyone a chance to read and post an answer. On a site like StackOverflow I wouldn't wait this long, but because of this site's low traffic at this point, I want to make sure enough people get a chance to participate.

• After some time goes by, I'll look at past questions. If one has a complete, useful answer (even if it's the only one) I'll accept it. If one of the answers is useful but seems to be missing a detail I mentioned in the question, I'll ask in a comment. If none of the answers seem correct or complete, I'll leave the question open in case someone comes along with a better answer later.

• If there isn't a complete answer to a question I'm very interested in or curious about, I'll start a bounty and ask for more details.

In the end, everyone approaches accepting answers in a slightly different way. But it's definitely something we should encourage people to make sure to do, to encourage good, complete answers and give credit for time invested helping the asker.

Answer 2

I totally agree with you that it's a negative practice, no doubt on that. Unfortunately, we can't really force the OP to accept and we even can't accept it by ourselves, of course.

So what can we do? My proposal is to make the translation requests requirements much stricter. At least, you'll see that only really good questions will be asked. And it's better for them to be still opened, than thousands more of lower quality.

On the Chinese SE we proposed a single but definite requirement for translation requests, I'll quote it verbatim:

Can I request a translation?

We do not forbid translation requests but note that this is not a translation service, so we will close any translation request that does not show research effort. Make sure you clearly show it in your question.

That's the key in my opinion, the research effort. I prefer to answer some (apparently or not) simple translation question if I see there clearly was research, rather than answer a more difficult question without any effort shown in it. Questions that can be easily answered by yourself must be closed. If you find the answer, but it's not clear, then I'd say it'd be ok to ask, but you/we/everyone is not here to work as translators for others.