Why does Stack Overflow using voting for questions?
Voting on a question means something like
- "Hey, that's interesting, I'd also like to know the answer (+1)", or
- "Reading this question turned out to be an utter waste of time (-1)".
(On meta, things work a bit different, however, e.g. votes on feature-request indicate "want, too (+1)" or "please don't, that would annoy me (-1)", or bugs are confirmed (+1))
My main reasons for upvoting the questions related to my interest are:
(1). I would like to know how to do this if I don't know the answer.
(2). If I know the answer but somebody proposed a better way of doing it then I would upvote both as it helped me learning something new.
Main reasons for downvoting:
(1). User has put no efforts in solving the question. Just looking for somebody else to solve their problem.
(2). Asking the same question again and again and again even after the previous one was closed.
Voting allows users to identify which answers the community feels are most correct. It also helps build reputation. The assumption is that answers by users with a higher reputation might be more reliable, presumably because they are more experienced. This is not always the case, but it generally follows.
In SOFU and the SE 2.0 sites, voting on questions expresses the importance of the question for the community: there are other people than the OP that are interested in the answers.
Votes on questions in SO only reward the OP with 5 rep points and therefore do contribute to the reputation of the person less than votes on answers. This was done to make it harder to build a reputation just asking questions. Yet asking a good question, provoking nice answers, is vital for a Q&A site. Therefore the rewards.
Votes on questions in meta have a slightly different catch for feature requests and bugs. People express their interest for the implementation of the feature or testify to the bugs existence.