What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 134 Stack Exchange communities.

This question already has an answer here:

The question here: Is it possible to forward declare a static array was asked a long time ago (2009) and earlier this year a solution was given by the user D.A. However, the user (now inactive) that posted the question has not selected it (or any other) as the answer.

After looking at "similar questions" while typing this out (e.g. Allow users to mark "accepted" answer on someone else's question?) , I see that this discussion has come up before. The counter argument is, "That's what voting is for." Back to the c++ question, the first answer, with 4 votes, is that c++ does not permit the desired behavior. Down below (at zero votes until I voted for it) is D.A.'s answer. Clearly, that counter argument only makes sense regarding answers that happened to be submitted at around the same time period, and have had an equal opportunity to receive votes. I would have completely missed the answer if I hadn't spent the extra time to look at the likely-wrong answers. If the site is to be helpful to users (and not just zombie accounts) then it makes sense for these kinds of answers to be marked as accepted so that they will be seen. Voting, in this scenario, is inadequate.

share|improve this question

marked as duplicate by Emrakul, U2744 SNOWFLAKE Aug 6 '13 at 19:14

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.

No, that thread does NOT answer my question!!! –  Brent Aug 6 '13 at 18:19
the attitude that anything other than the top voted answer (especially when nothing has been accepted) is "likely-wrong" is what is inadequate and should change. –  Kate Gregory Aug 6 '13 at 18:20
The top answer has 4 upvotes. It's not like it has 200. If there's a radically better answer posted, even years later, it's not that hard at all for it to get more than 4 votes and be at the top of the list. Voting most certainly is capable of handling this case. –  Servy Aug 6 '13 at 18:24
Who is going to vote one way for voting and annother for accept voting? –  Richard Tingle Aug 6 '13 at 18:26
@Aaron yes but thats you disagreeing with the OP's accept vote, im saying who is going to disagree with their own accept vote. If not then you might as well just have regular voting –  Richard Tingle Aug 6 '13 at 18:30
@RichardTingle well, an OP may find multiple answers helpful, but they can only accept one (the "best" answer, or the one that helped them most). So what I see often is that an OP will accept that answer, and up-vote the others that were good or that they found helpful. What Brent is trying to address is the case where the OP has completely disappeared, and never accepted (and probably nevery up-voted) any answer. –  Aaron Bertrand Aug 6 '13 at 18:33
We need a concrete proposal as to your solution (moderator accept, community vote to accept etc) otherwise we're discussing our own assumptions –  Richard Tingle Aug 6 '13 at 18:53
@Richard yes, agreed. This is currently just "there's this problem with this one question, and I want the user to accept it!" rather than "there is a widespread problem, here's what it is, and here's how I think we should fix it." –  Aaron Bertrand Aug 6 '13 at 18:55
@Richard as I mentioned in the one answer, bringing more eyes to it so that the community can decide might work. And Aaron brought up a good point - bounty does something similar. The way that it doesn't work so well is that the question may already be answered, and bounty is for questions that aren't. Maybe users with enough rep could get the question to be shown again, and then a fresh round of votes determines the accepted answer. –  Brent Aug 6 '13 at 18:56
@Brent bounties are not just for questions that aren't answered. Just the other day I gave a 500 rep bounty to an accepted answer from months ago. Perhaps you should learn more about bounties before dismissing them... –  Aaron Bertrand Aug 6 '13 at 18:59
Anyway, I still disagree with the entire premise, even if you could unify what you're expecting here (moderator involvement, collective, etc). The checkmark means that the user accepted the answer as the best answer for them. Checking that box for them is lying, regardless of who is saying that they can make that judgment call. The best answer is going to be the one that's up-voted the most. I often see accepted answers with a negative score, usually because the OP didn't realize how bad the answer was. –  Aaron Bertrand Aug 6 '13 at 19:02
@Brent, it is your privilege to choose the answer you're awarding the bounty to in the first place. –  Frédéric Hamidi Aug 6 '13 at 19:05
@Brent well again I think you should understand how bounties work before you go creating one. You've stated several misconceptions and false assumptions about them already. –  Aaron Bertrand Aug 6 '13 at 19:05
Also, the bounty will reward the selected answer, and should get more eyeballs on the question, but there is no guarantee that it will get more up-votes. So don't create a bounty, reward it, and then say "see, look, the bounty system is broken too!" Other people have to see the new answer and agree with you that it is better than the other answer. Another way to get more eyeballs on the new answer is to comment on the question (or the current "best" answer), and indicate that a new answer is better (and why). –  Aaron Bertrand Aug 6 '13 at 19:13

1 Answer 1

If you're suggesting that users should single-handedly be able to mark the accepted answer for someone else's question, that is extremely open to abuse. If you're suggesting more of a community based, multiple vote system, we already have that. It's called voting.

share|improve this answer
I was thinking more like moderator involvement. –  Brent Aug 6 '13 at 18:31
@Brent moderators aren't qualified to judge answers. Moderators aren't chosen for their subject level knowledge. Only the community is qualified to make that judgement and thats voting –  Richard Tingle Aug 6 '13 at 18:32
Good point. I'm sure someone more clever than me could figure out something to enable the users to re-evaluate old questions and perhaps collectively select the correct answer. Would take some work on SOs part though. –  Brent Aug 6 '13 at 18:35
@Brent that simply won't scale. Even as a moderator on dba.SE, I am quite comfortable with SQL Server, but I wouldn't be able to choose the best Oracle answer (never mind even know for sure if any of the answers actually answered the question or helped the OP). I would be relying on input from experts in the community to determine what the best answer would be, and they can already do that - by voting. –  Aaron Bertrand Aug 6 '13 at 18:36
@Brent "collectively select the correct answer..." But isn't that the same as voting? –  Asad Aug 6 '13 at 18:36
It's really not. –  Brent Aug 6 '13 at 18:36
@Brent if you are really passionate about a particular question that is four years old, has a new answer, and think the new answer deserves more attention, give up some of your rep and offer a bounty. –  Aaron Bertrand Aug 6 '13 at 18:37
@Brent to you, it's not. To everyone else who disagrees with you, it is. –  Aaron Bertrand Aug 6 '13 at 18:37
Bounty still doesn't get the user to accept an answer, no? –  Brent Aug 6 '13 at 18:38
Voting is different because visibility is important. Accepted answers are seen at the top, right away. –  Brent Aug 6 '13 at 18:38
@Brent, why are you so keen on having the answer accepted? An answer being accepted only means it was the most useful to the questioner, not necessarily to the rest of the world. –  Frédéric Hamidi Aug 6 '13 at 18:39
@Brent no, but if your theory is right that the new answer is so much better than the currently most-voted answer, new eyeballs should drive up-votes to help push that new, so-much-more-obviously-correct answer to the top (when people order by votes). –  Aaron Bertrand Aug 6 '13 at 18:39
@Brent How would it differ? People cast votes to indicate the accuracy and quality of answers. The answer that community consensus dictates is the best will have the highest score. If your objection is merely that the answer the community selects should be highest in the sort order, that is a different argument altogether. Changing the visibility of highly voted vs accepted answers would not require a parallel voting system. –  Asad Aug 6 '13 at 18:39
@Brent could you find 10 more questions - out of the millions on Stack Overflow - where this is a problem? –  Aaron Bertrand Aug 6 '13 at 18:48

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