The tooltip for up voting a meta question suggests that it is appropriate when the question is clear, shows prior research, and is useful. A down vote is appropriate when the question the opposite is true. I believe these tooltips are common throughout meta and parent sites.

If there's a meta question that is poorly researched, or unclear, or is not useful, then this is straightforward. However, there is a practice (presumably originating on meta.stackoverflow.com) where these are not the reason for up/down votes; rather they indicate agreement or disagreement with a suggestion in the meta post. I presume this is because meta.stackoverflow.com users are too lazy to add a "Yes" or "No" answer or vote such an answer up or down.

Based on my own experience (I've used Stack Overflow for 3 and a half years, but had limited meta experience before this site) and comments I've seen others make, this is especially confusing for new meta users, who are ones we want to be encouraging.

I think we need to have a policy about this so that everyone either knows what to do or can figure it out by reading something on our own meta site.

  • 2
    The downvote here is another example. Shouldn't have asked the question? Asked it poorly? Shouldn't tell people what voting on a question means and should leave them to guess?
    – Tony Meyer
    Commented Feb 3, 2012 at 5:40
  • 4 years later and it is still confusing to new users!
    – Skooba
    Commented Jun 14, 2016 at 14:00

2 Answers 2


If we stick with the vote up/down questions to indicate agreement/disagreement system, then I think we should more carefully format discussion questions so that it's clearer what's being voted on.

For example, in this question I ask "should we keep the blog alive" and talk about whether it should be kept going or not. It's not clear what an up vote means under the "agreement"/"disagreement" system, so it should be edited so that it's proposing something specific (e.g. "we should shut the blog down").

Questions that aren't clear can be edited so that they are. The volume of questions is low enough that this is easily manageable.

  • I upvooted the blog question because I wanted to keep it alive. So I was using the votes as a yay/nay vote Commented Feb 3, 2012 at 5:01

The practice of using a question vote as a poll is confusing. I'm often unsure about how I should vote on a meta question - if it's something I think is useful (i.e. it should be discussed), and it's well written and presented, but I disagree with the suggestion, do I vote up (great question!) or down (terrible idea!)?

If I'd like to know what other people think of an idea (e.g. on a question I've asked), I have the same problem in reverse - did people vote it up because I asked something that should be asked and did it well, or did they agree with the suggestion? Did people vote it down because I asked it badly, or because they disagree with the idea?

I think we should break away from the bad habits of meta.stackoverflow.com, and say that if you agree with a suggestion, you vote for an answer that says so (and add one if there isn't one). If you vote for a question, you're saying that it's something that many people should read.

Answers are answers, questions are questions.

  • 3
    I agree in concept, but I seriously doubt the practicality of this. As was noted, The Powers would likely prefer SE conformity over good design. I think this MSO practice is VERY confusing for new users, but MSO was never very welcoming to new users IMHO. Commented Feb 3, 2012 at 7:05

You must log in to answer this question.

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