What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 127 Stack Exchange communities.

When a question is closed as a duplicate of another, should the other question's score be affected?

I'm thinking this would be a good secondary mechanism for promoting the visibility of questions which get asked (i.e. duplicated) a lot.

The relationship doesn't necessarily have to be linear. For example, there could be a staggered threshold, like for 1 duplicate = +1, for 5 duplicates += 1, for 20 duplicates += 1 or whatever is empirically sound.

share|improve this question
2  
No. I can't see any benefit from adding to the reputation of the original question author. Maybe add another meta data on the question page showing how many questions are duplicate of that one, but it got nothing to do with reputation. –  Shadow Wizard Sep 12 '12 at 6:31
    
No no, not the author's reputation; the question's score, and specifically the innocent target question which was duplicated. Thanks for the feedback; updated my question's title. Hopefully it's clearer now. –  tripleee Sep 12 '12 at 6:41
    
As 10K member I'm very surprised you still don't know how reputation work. Question score is reputation, where 1 positive score is +5 reputation to the question author and 1 negative score is -2 reputation. (For example this question of yours with score 3 gave you 15 reputation) –  Shadow Wizard Sep 12 '12 at 6:43
    
I know how that works, but thanks for reminding me. What I am getting at is increasing the visibility of questions which ought to be popular -- maybe question score is not the best mechanism for that, but then what is? I do not consider it a problem that the asker of a question also gets an increased score, but if you do, perhaps you can explain why it is not desirable? –  tripleee Sep 12 '12 at 6:47
4  
Having other questions closed as duplicates of your own does not mean it's a good question. Bad questions can also get duplicates. This alone is reason enough in my opinion. –  Shadow Wizard Sep 12 '12 at 6:59
add comment

3 Answers

up vote 2 down vote accepted

This is counterproductive to the goals of Stack Overflow. If you look at other gamification strategies on the site, such as the reputation system and badges, you'll notice that they're carefully crafted in such a manner that desired behaviors are encouraged and rewarded while undesired behaviors are discouraged.

We don't want to necessarily discourage duplicates, since it helps improve SEO and expands the possible keywords that can lead to an answer, but we don't necessarily want to encourage duplicates either, as this is something that could be massively abused and easily get out of hand.

The first step before posting a question is to search, search, search, and if you happen to find the answer to your question, then great. Problem solved! But if not, or if the existing answers don't really fit your situation, then ask away.

However, we don't want a carrot hanging at the end of a stick that's going to encourage people to post without searching. This would be detrimental to a site as large as Stack Overflow and would clutter it up with a lot of useless, exact duplicates that have no value. If a question and its answers are great, they'll get upvoted, and linking them as duplicates actually increases the chances that a user landing on post A may go to post B if the answers on post A don't solve his or her problem.

Additionally, votes are the voice of the community, and artificially increasing votes on a post don't reflect the opinion of the community. I believe this is what Sha Wiz Dow Ard was alluding to in the comments.

share|improve this answer
    
Thanks for a useful answer! My rationale for this suggestion is precisely to discourage users from posting duplicates. My line of reasoning is that if a "canonical" answer gets sufficient score, it will fare better in search engines (including, but not limited to, the site's own, and Google). I don't see how this in any way can be construed as hanging a carrot on a stick. Could you please explain? Of course, I do realize that a lot of duplicates are posted by users who do not search at all, but obviously, this proposal does not address that scenario. –  tripleee Sep 12 '12 at 9:01
    
People could game the system, either by intentionally trying to get their duplicate closed as the canonical artificially, or by closing questions that are borderline, or encouraging other users to close borderline questions that otherwise wouldn't be closed. Not saying it will happen a lot, but someone will try to abuse it in some manner, perhaps in ways we haven't thought of yet. :) –  jmort253 Sep 12 '12 at 14:54
add comment

This feature request would make it way too difficult for users to understand how a certain question score was reached. Now, it's simply (up-down)votes by the community which is clear to everyone. Let's keep it that way.

share|improve this answer
add comment

Most often, when a question is detected as a duplicate, there are many possible choices because it was asked many times.

Your proposal would give too much importance to the selection of the best duplicate, and thus need a process which is totally incompatible with the time we want to give to a question that has, by nature, a somewhat low worth.

And those frequent questions aren't usually the ones whose answerer has the greatest merit as the most basic a question is, the most frequently it is asked. I'm yet a little ashamed by the fact the questions that gave me the greatest reputation would have been answered by somebody's else if I had been a few seconds late, I don't want them to gain more upvotes by an indirect mechanism.

So I think this would be unfair and heavy.

share|improve this answer
add comment

You must log in to answer this question.

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