# Is there some reason not to up-vote a question while voting to close it as a duplicate?

Once a couple of years ago, I found someone to be under the impression (which seemed unreasonable to me) that one should not up-vote a question that one votes to close. I don't know why it would be considered unreasonable to up-vote a question while also voting to close it as a duplicate. Up-voting means its a question that is worth having on the site. Closing as a duplicate means one should look elsewhere for the answer.

Now I find this same idea obliquely suggested in a comment under a question.

Is there any official position saying that that is an error? I think that would be useful.

• One possible reason for not upvoting: assuming that the poster has not tried searching the site... – Aryabhata Jul 23 '14 at 1:40
• @Aryabhata: It may be a reason, but I disagree with the assumption. Often, the only way to find a similar question is for a question to be marked as a duplicate by someone who has seen a similar question. Generally, searching for math questions is not easy. – robjohn Jul 23 '14 at 6:02
• "Up-voting means its a question that is worth having on the site." But it's already on the site, if it's a duplicate; it's not adding any value, having it on the site twice, is it? – Gerry Myerson Jul 23 '14 at 7:26
• @GerryMyerson : So vote to close it. I already said that. – Michael Hardy Jul 23 '14 at 18:24
• Yes, but you also said that you didn't know why it would be considered unreasonable to upvote a question while also voting to close it as a duplicate. I'm telling you why it's unreasonable. It's unreasonable because, as you say, upvoting means it's worth having on the site, but a duplicate is not worth having on the site. OK? – Gerry Myerson Jul 24 '14 at 0:16
• @robjohn: I agree with you, it is hard to do a search. To do a good search you already need to be aware of the latex capabilities, some common ways to write the same math differently etc. Too much to expect from newer users. Uniquation helps a bit in that regard (see this: meta.math.stackexchange.com/a/1394/1102) – Aryabhata Jul 24 '14 at 0:16
• @GerryMyerson : It's not worth having a duplicate of a question, but it's worth having the question. Anyway, the particular question that inspired this present discussion was about whether there are infinitely many prime numbers. I did a search for that and didn't find it, and I think I'm better at choosing search terms than the questioner was. – Michael Hardy Jul 24 '14 at 2:13
• I agree that it's hard to search this site, and I don't hold it against anyone who unknowingly posts a question already present. But here we're talking about upvoting a question when the voter already knows it's a duplicate. – Gerry Myerson Jul 24 '14 at 3:15
• @GerryMyerson : Correct: The voter knows it's a duplicate, and up-votes it and votes to close it because it's a duplicate. – Michael Hardy Jul 24 '14 at 4:51
• The voter knows it's a duplicate, so the voter knows it's not worth having on the site. But the voter upvotes it, signifying that it is worth having on the site. I smell a contradiction. – Gerry Myerson Jul 24 '14 at 5:37
• @GerryMyerson : The voter knows that it's a duplicate and that it is worth having on the site, but not as a duplicate. – Michael Hardy Jul 24 '14 at 16:46
• I rest my case. – Gerry Myerson Jul 25 '14 at 0:00
• . . . and so the voter up-votes it and votes to close it. – Michael Hardy Jul 25 '14 at 0:01

The SE position on voting is that users are free to vote as they want, as long as they do not target one another. As a corollary:

• If you want to up-vote a duplicate post, then up-vote it
• If you want to down-vote a duplicate post, then down-vote it
• If you don't want to do either thing, then don't.

Comments suggesting that someone should or should not vote up/down are pure noise.

• In comments a few weeks ago, someone said it was not clear to him what an up-vote on a question means. I told him I take it to mean the question is worth having on this site and worth posting an answer to, because it helps people learn. He reported that that short comment of mind convinced him that that's the rule that should guide him. Was my comment "pure noise"? – Michael Hardy Jul 23 '14 at 0:37
• @MichaelHardy On the main site, it's still noise; something I would delete after the message was received. – user147263 Jul 23 '14 at 0:57
• But if one user found it useful, others might also find it useful. – Michael Hardy Jul 23 '14 at 3:22
• @MichaelHardy Someone might find a roasted turkey recipe useful, but it still does not belong on Math.SE. – user147263 Jul 23 '14 at 3:24
• Something like that does belong on the site. It helps people use the site for its intended purpose. – Michael Hardy Jul 23 '14 at 18:25
• @MichaelHardy "How to use the site" is meta content (it is about the site itself). Therefore, it belongs on $\underline{\phantom{meta}}$. – user147263 Jul 23 '14 at 18:27
• ^ I disagree. Long "meta-" discussions belong on meta. Short comments that can improve the way people use the site might never be seen if they're on meta. – Michael Hardy Jul 24 '14 at 2:14

Posting another answer to separate my personal position from generic SE advice.

1. I will never upvote a question if I know it to be a duplicate.
2. I will downvote an otherwise reasonable duplicate if I think the asker did not put enough effort into search.

My reason lies in incentives. If duplicates are never downvoted, users have zero incentive to search before asking. If they are habitually upvoted, there is a negative incentive to search: reputation is given for asking a duplicate, not given for finding an answer via search.

• I agree with point 2, but only in the cases that it is completely obvious that no search was done; but I find that fairly rarely, considering how difficult it tends to be to search through mathy-titles. I will sometimes upvote a duplicate if it's much better asked than the original; my upvote is then meant to be a compliment for asking a particularly good question. – user61527 Jul 26 '14 at 4:36