I understand why we mark questions as duplicates and close them-- it makes sense.

However, I don't think downvotes and duplicates should go hand in hand, which is how the system currently works.

Currently, if someone asks a valid and intriguing question useful to many people, but which has been asked 10x before, they will immediately get 1-4 downvotes. This has happened to me and I've seen it happen to other people.

This is really unfortunate, because often these questions are being asked by new people who are new to programming and new to StackOverflow. They have no idea that asking the difference between the JRE and JVM and JDK is an old question. They're new. Are there existing resources that answer their question? Of course there are, and that's why we mark it as duplicate-- to point them towards those resources.

But downvotes don't convey "hey, good question, but this has been asked before." They convey "This is a stupid question or badly asked. We're not going to waste time answering this." It makes people feel bad and it introduces an atmosphere of negativity that needn't be there-- simply marking a question as duplicate is effective, meaningful, and useful.

But how can we dissociate these two actions?

My best proposition is that marking something as a duplicate should remove its downvotes. If it's a duplicate of a question voted 6, a question we've communally deemed "good," then why does this new person deserve -4 for wondering the same thing?

I am looking for other ideas of how these things can be separated from each other, as I am not convinced this is the best way and certainly not the only way.

  • 18
    Implicit in the posting of duplicate questions is the assumption that the poster didn't do sufficient research to discover the duplicate before posting. If the question is a duplicate, why repeat it? The validity, intrigue and usefulness is already captured in the original post. – user102937 Nov 27 '19 at 2:41
  • 1
    If this is referring to stackoverflow then it should be posted on stackoverflow-meta not here. – A Rogue Ant. Nov 27 '19 at 2:46
  • 1
    It applies to all SE sites. I thought that seemed fairly self-evident. – temporary_user_name Nov 27 '19 at 2:46
  • 1
    @temporary_user_name Not particularly, as you stated the name of only one site - stackoverflow. I genuinely don't know if you've any connection to any other site. – A Rogue Ant. Nov 27 '19 at 2:49
  • But it seems clear that the principle could occur on any site, no? Of a duplicate question being downvoted? I don't understand why you're trying to imply that that needs to be explicitly stated. – temporary_user_name Nov 27 '19 at 2:50
  • @temporary_user_name But you specifically (and only) stated stackoverflow. Why would anyone assume you meant all sites? – A Rogue Ant. Nov 27 '19 at 2:52
  • Just seems self-evident that it goes for all sites regardless of which specific ones I use as examples, given that it's a platform-wide mechanic. I can't imagine why anyone would think it only applied to one site when they're all identical. – temporary_user_name Nov 27 '19 at 2:54
  • @temporary_user_name We're (by which I mean myself) particularly sensitive to SO's pre-eminence over the other sites concerns regarding The Loop being only about stackoverflow. (or at least not being explicitly about the whole network). – A Rogue Ant. Nov 27 '19 at 3:01
  • Well, apologies, I haven't really paid any attention to their blog posts / corporate developments. – temporary_user_name Nov 27 '19 at 3:02
  • 3
    What about upvotes? There are cases where a question is upvoted, presumably by people who aren't aware that the question is a duplicate, before it's closed as such. Should such upvotes be removed as well? – DK Bose Nov 27 '19 at 7:05
  • 2
    Would you also be in favor of removing from the mouseover text, "This question does not show any research effort"? – DK Bose Nov 27 '19 at 7:08

One key element here: when writing a question on this network, you are, in general, expected to do prior research.

A duplicate often means: you didn't do that. And the more obvious "absence of prior research" can be found, the more people are likely to downvote.

The real issue is (imho) twofold:

  • The "new question" wizard logic should make it as simple as possible to identify existing questions for the same problem. Last time I checked, it wasn't even close. It improved over time, but still: long way to go. When google can find a better duplicate from a "new" question title alone, why can't "our" search deliver that?!
  • Yes, the experienced users could be a bit more reluctant on downvotes. When I encounter a "dup" that is well written for the SO java tag, I will simply close as such, but without a downvote. Unfortunately, when I come back 30 minutes later, there might be 2, 3 downvoted already.

And looking here for example, the "majority" consensus is: no need to downvote duplicates.

  • 7
    Google is a better identifier of duplicate questions. – user102937 Nov 27 '19 at 2:52
  • 1
    I totally agree. That is the thing that I totally don't understand. Why can't they come up with better search?! – GhostCat Nov 27 '19 at 3:31

Not very much.

Votes are confidential, and only staff (or the voter if there was an edit) can undo a vote. Most people who do...would be disinclined to.

Unfortunately, if you wanted to fix it, you would need to bring about a widespread, pretty monumental change in voting culture, and technically, not finding a dupe yourself could be counted as insufficient research

You must log in to answer this question.

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