I don't know for sure, but at least for me my chances of checking a question drop significantly when it already has an answer (and will continue to drop as the question gets more answers). The problem is that this leads to some questions not getting good answers because someone who actually knows the answer won't even look at it because it seems like it has already been answered.

I think it would make sense for a question to either appear unanswered (probably too misleading) or at least make it clear that all answers to the question have 0 or perhaps -1 or less votes. This will help to increase the chances of someone else viewing the question and providing a better or real answers.

There are times when I ask a question and I check back and it has an answer, but the user did not understand or properly read the question. I notice that these questions rarely get other answers, but I can't signal anyone that the answer I got is no good and I need more help. Duplicating the question is no good either.

  • 3
    If you routinely get unhelpful answers then there's a problem with the way you ask your questions. No amount of tinkering with the system is going to fix that. Jun 14, 2011 at 14:40
  • 2
    This implies that all responsibility for a question is on the asker and anyone who answers is exempt. If that's the case, why are we able to downvote answers? Jun 14, 2011 at 14:46

5 Answers 5


The questions page already differentiates between those with an accepted answer and those without one, thus the most important mark is already implemented.

Furthermore, if an answer is bad/incomplete, it will be eventually downvoted and peer pressure may force the user either to delete the answer or to improve it. A comment stating why the post is not answering the question will help greatly in that respect.

Lastly, think also what is your responsibility for the lack of a good answer: is the question clear enough? How did it mislead the user trying to answer it? Finding the weakness and edit your question will help in two ways: 1) a clear question attracts better answers; 2) the edit will bump the question to the top of the question list.

  • 3
    I agree with you in some respects, but I when I look at the unanswered questions page, I still have a much lower chance to check a question that has already been answered even if it has not already been accepted. There is a difference between a bad answer and an answer that has just not been accepted yet, but there is no way to tell this before you even see the question. You are right that a bad answer may come from a bad question, but this is not always the case. Instead you are keeping people who may have better answers out. In this argument, editing the question does nothing. Jun 14, 2011 at 14:09
  • 1
    Not everyone avoids questions that already have an answer. I, for example, don't. I may avoid questions with 8 answers, but I know that one answer may not mean anything. Bumping the question (with a useful edit) will help.
    – Aleadam
    Jun 14, 2011 at 14:21
  • BTW: you already have three answers here. That proves the point that people still answer when there's already an answer.
    – Aleadam
    Jun 14, 2011 at 14:22
  • I never said that people don't answer when there's already an answer, but this is also a discussion rather than a question so there is far less chance of redundancy for answers. There may not be a correlation between answers and avoiding questions, but I think there is one. I do it. Jun 14, 2011 at 14:38
  • If that's a common behavior, then I think we should do something to change that and promote more answers, rather than stimulating a not useful status quo
    – Aleadam
    Jun 14, 2011 at 14:48
  • @Aleadam: Don't forget that this is Meta here. On the SO homepage, most of the questions shown have 0 answers - I'm not sure, but this might have some impact. Jun 14, 2011 at 15:53
  • @Hendrik are you replying about the answer or about my comment that there are already three answers here? I agree with you if it's for the latter, but I do think the options in the answer are all valid for SO.
    – Aleadam
    Jun 14, 2011 at 17:56
  • @Aleadam: Sorry, I should have said above that I'm talking about your comment "BTW: you already have three answers here ..." Jun 14, 2011 at 18:40
  • @Hendrik indeed I did not plan to put a lot of weight in that phrase, and that's why it was just a comment and not part of the answer (mine was not the first answer). Anyhow, by looking at the first page in SO, I see a lot of zero-answer questions and then the rest divided between questions with one and with many answers. Single-answer questions are not the majority in there, which in itself validates the fact that people will answer questions that have already an answer.
    – Aleadam
    Jun 15, 2011 at 4:41
  • @Aleadam: I also didn't want to put too much weight into that. As I wrote, I'm not sure how big the impact is, but I also don't quite follow your "evidence". It would need a bit more to make a proper statistic; for starters I have no idea what answer distribution to expect. Jun 15, 2011 at 6:10

I think the real problem might be this:

my chances of checking a question drop significantly when it already has an answer

I can't speak for everyone here, but I don't remember ever ignoring a question just because it had an answer posted. An accepted answer, sure. Ten answers, maybe. But one answer? Without even checking to see if I could offer a better one? I doubt that would stop me unless I wasn't very confident in the quality of my answers.

If anything, I've been a little disturbed by how often I read a question with three answers already, and none are all that helpful, thoughtful, or even correct. Usually, I read through the other answers, and if I feel like I can do better, I try to contribute. If I can't do better, there's a good chance I'll learn something.

I'm really, truly not trying to be insulting or condescending here, but you may find that the better your average answer becomes, the less worried you'll be about competition. A lot of those first answers are slapdash FGITW-type things anyway.

  • 1
    Personally, I definitely favor unanswered questions, but I don't completely ignore answered ones. Jun 14, 2011 at 15:51
  • I don't think you come off as condescending or insulting, but I do think that thinking of someone else's behavior as a problem may be misguided. Remember that people don't have to use things a certain way even if they were specifically designed to be that way. That doesn't mean the designer has to change his design to suit these people he thinks are wrong, but it doesn't hurt for them to ask. I'd also like to point out that I don't ignore questions with answers (even many answers if it looks interesting), just that I will look at an unanswered 100% of the time. 1 answer, maybe 80, 2 60.. Jun 14, 2011 at 16:17
  • 1
    @tandu - I'm not saying you have to use the site in a certain way, far from it. I'm just saying that it's (unfortunately) common for the first couple answers to leave lots of room for improvement. My point wasn't that your behavior is a problem; I guess what I'm saying is: Why should a couple of answers scare you off? There's a good chance you could do better. Jun 14, 2011 at 16:30

One can start a bounty if a question doesn't has a good answer.

  • 3
    This is a good system if the answer is not sufficient for you, but I don't think it's fair that you should have to sacrifice rep if an answer was provided that should have not been given in the first place due to being misleading, wrong, etc. and it's preventing a good answer that would have come for free otherwise. Jun 14, 2011 at 13:06
  • 7
    Bounties are nice for those of us with high rep, but this isn't a good solution for people without a lot of rep to spare. A bounty should be something used on a very difficult question, which has been open for a long time (there is a minimum time for even starting a bounty). It shouldn't be something you have to use on a regular basis.
    – Kibbee
    Jun 14, 2011 at 13:07

I'm not convinced this would actually have that much impact.

From what I've seen, there are four main reasons that questions attract answers:

  • The question has just been posted (this seems to be the main criteria), with most answers being posted quickly after a question has been asked.
  • Somebody edits / posts an answer to the question (bumping it to the front page).
  • The question has a bounty assigned to it.
  • The question doesn't have any answers yet and somebody comes across it (possibly through the unanswered tab).

Your proposed change would have the most impact on people that browse the unanswered tab, but consider this question. You answered it and currently have a score of -2, however it has been accepted, the community score hasn't really changed / reflected the usefulness of the answer to the person asking the question.

The number of votes an answer has had can be an indication of the quality of the answer. However, it's also a reflection on the number of people that have viewed the post + voted for answers. A lot of people don't seem to vote that much and certain tags don't get that many views, so you can end up with several answers that actually answer the question, none of which have been upvoted...


One of the things that confuses my is why given your relatively low vote counts (yours is 91 votes (up to 6 negative votes) on answers for 304 answers given by you) you believe this would significantly alter the number of answers you get to your questions, or the n umber of questions you look at on the unanswered tab.

Looking at your questions, I can't see any negative scored answers on the questions you've asked, so presumably any answers that were wrong have been deleted, or rated by the community to be useful even if they aren't solving your problem. Of your 3 questions that don't have accepted answers none of them would be helped by the change you've proposed (one has an answer with a score of 1, another has a score of 4 and the third question doesn't have any answers).

  • That question seems to have two accepted answers (bug?) and I don't think my answer is any good anyway. You are right that it would have the most impact on the unanswered tab. That's the tab I use almost exclusively. Your argument that votes on answers don't reflect their usefulness is spurious. Sometimes they don't, but I believe they are designed so that they should (if not then what?). I agree that answers with no upvotes may be great, but at least an answer with a downvote (especially two) has a high probability of being poor quality, no? Jun 14, 2011 at 14:44
  • @tandu: A downvoted answer has a reasonable chance of being poor quality, I don't know if I would say high. People downvote for various reasons, including 'I don't think you should have answered this question' / 'The op wants to do XXXX, but you should have told them to do YYYY instead'.
    – forsvarir
    Jun 14, 2011 at 14:51
  • @tandu: I didn't say votes don't reflect usefulness. I said that wasn't the only thing they reflected. Votes on answers reflect the relative usefulness of an answer within a particular question, to the community. If the question has been viewed by 1000 people and an answer has two votes, I'd say it's less useful than if the question has been viewed by 5 people and has the same number of votes.
    – forsvarir
    Jun 14, 2011 at 14:55
  • Whoa...off topic, but do you know why the question you referenced has two accepted answers? How can that even happen? Is it because the accepted one has a negative score? Jun 14, 2011 at 16:01
  • @Justin Morgan: I flagged the question, with this possibly related meta question meta.stackexchange.com/questions/76952/….
    – forsvarir
    Jun 14, 2011 at 17:49

But this is already so; we define "unanswered" as "question with no upvoted answers", not "question with no answers".


You must log in to answer this question.

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