Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 158 Stack Exchange communities.

What is meta?
Here's how it works:
  1. Any Stack Exchange user can ask a question
  2. The community provides support, votes on ideas, and reports bugs
  3. Your voice helps shape the way Stack Exchange operates

I don't think closing as duplicate is functioning in the way it is supposed to. I don't think closing is the right way. There are certain kinds of questions that are repeatedly asked, and even if they are closed, it is often too late in the sense that answers are already given. Furthuremore, the number of duplicate questions keep growing, and over the time, it gets more and more difficult to extract non-duplicate answers from the floods of answers distributed among the duplicate questions.

Why not have them automatically merged after certain amount of time after closing (this time is for the possibility of a question being reverted as non-duplicate)?

I think that a question judged to be a duplicate of another is a good indication that it is a common question. Automatically changing it into a wiki might be another way.

share|improve this question
What would happen to questions that have 5+ answers every time they're asked because in reality they're bike-sheds that anyone can, and will, answer? The net result would be that the remaining question (the one you merge into) would grow by 5 new answers that would basically duplicate existing answers there every time you close such a question. Not knowing the full process of closing the question (beyond it being closed), is there a "close but review" process so that moderators can migrate good answers to the remaining question, or are they just lost? – Lasse V. Karlsen May 23 '11 at 9:42
@Lasse That is a good point. Certainly something like you suggest will be necessary. – sawa May 23 '11 at 10:28
@Lasse We cannot selectively migrate answers. We can only strike the whole set at once. That said, if you think there are answers worth merging, then your "close and review" process is accomplished by closing it, then flagging it for moderator attention. We will review the question ourselves before deciding whether or not it will be wise to merge it. – Grace Note May 23 '11 at 13:23
@Grace Yup, that's what I assumed. – Lasse V. Karlsen May 23 '11 at 13:29

I can think of three things that already solve this problem, without the need for adding additional complexity to either the site's user interface or the source code:

  1. We can simply continue to close duplicate questions as quickly as possible, before a bunch of incomplete, bike-shedding, and/or just plain duplicate answers have a chance to be posted. Judging by the number of questions posted here on Meta with users complaining how quickly their question was just closed, it seems to me that this system is working as intended.

    Sure, some questions don't get closed as quickly as others. But those are generally the ones in the less-popular tags that get fewer views (and thus have fewer chances for close votes to be cast). But in the less-active tags, it seems that your cited problem of rampant duplication would be much less for reasons of scale alone.

    (And if you want to help ensure that duplicates are sought out and closed even more quickly and effectively than they already are, support Pekka's proposal to provide reputation bonuses for it.)

  2. Duplicate questions whose answers contain nothing of added value compared to the "master" question can (and often are) deleted shortly after being closed. All 20k+ users can vote to delete immediately, 10k+ users can vote to delete after 2 days, and moderators can delete questions with an immediate, binding vote.

    Again, I think this is already happening, judging by the handful of people that have complained here on Meta about questions being deleted.

  3. If you ever come across a duplicate question that has some really good answers that provide information not already provided in the answers to the duplicate question, you can (and should) flag that question for moderator attention and ask them to merge the two questions. Moderators already have this capability, and the necessary discretion to apply it only when appropriate. This is a great option for when deletion would otherwise be inappropriate.

And finally, as a special bonus argument, consider that some duplication is actually a good thing:
Dr. Strangedupe: Or, How I Learned to Stop Worrying And Love Duplication

TL;DR: It seems to me that the tools already in place allow us to deal effectively with the bad duplicates, prevent their number from continuing to grow, and keep it easy to find the quality content that you need. And that the good duplicates are actually, well, a good thing.

share|improve this answer
I upvoted Pekka's proposal following your suggestion. I am with your suggestion, but there are certainly cases that sneak through them. For example, I was just reading a question on diff functionality in ruby, and according to my view, there are 17 duplicate question, none of which are voted as duplicate. This kind of situation is not rare. I can try for your second suggestion. I will try to see how much it works. – sawa May 23 '11 at 10:41
@sawa: Really? 17 duplicate questions? I don't hardly believe that. I've never seen such a thing in any of the tags that I follow, and they're the fairly popular ones. – Cody Gray May 23 '11 at 10:42

You must log in to answer this question.

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