What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 130 Stack Exchange communities.

I often see situations where a question is tagged as a duplicate because there is a different question whose answer also answers this question.

As an example, set attribute with javascript super method is marked as a duplicate of Why are my JS object properties being overwritten by other instances. While an explanation of javascript inheritance answers both of these questions, that does not mean that they are the same question.

This is important because the user asking about accessing javascript super classes would have no way of knowing that the a question about object properties being overwritten was the "same" question he had without reading the question in fairly serious depth. So he searches for his question, finds none that have a surface relevance to it, so he asks his question and then is told that it's a wrong question because a different question happened to have the same answer.

That seems really counter productive and harsh.

It has a further bad affect that at a later point someone asked the question Calling Superclass Method From Subclass which is a well posed formulation of the root question shared by those two topics, and it gets closed as a duplicate because of the two earlier questions.

The end result is that a well formed question that would be of use to a wide number of users is closed because earlier someone had asked a very localized question that happened to have the same answer. This seems very counter productive to building up stackoverflow as a useful repository of knowledge.

Would it be possible to refine our definition of duplicate to stress that the question has to be duplicate, not merely that an answer exists that would satisfy that question? Or alternatively have a different process to make "that's a new question, but this old answer works for it"?

share|improve this question
    
If an answer actually solves a question, I do not not see big problems to have a well formatted question marked as duplicate. If someone search this question he/she will be able to track the good answer, wont't they? –  Andre Silva Aug 8 '13 at 20:04
1  
Maybe the problem lay in having a question well formatted being marked as duplicate of a poorly question with an not satisfactory answer. –  Andre Silva Aug 8 '13 at 20:09
2  
That's certainly a component, if you have a thought out general interest question, you'd like a thought out general interest answer, not a "this is why the code you just showed doesn't work" answer that happens to cover the general topic. –  BostonJohn Aug 8 '13 at 20:11

1 Answer 1

Your question here seems to be based on a central assumption that is erroneous: namely that closing a question as a duplicate is a bad thing, or something to be avoided, presumably because of some message that it sends.

So he searches for his question, finds none that have a surface relevance to it, so he asks his question

That's exactly what he/you is/are supposed to do. All that we ask is that you make a best-effort attempt to search first, before asking the question. If you can't find a similar question that has already been asked, then you should ask the question yourself. No one did anything wrong here.

Often, the person who is experiencing a problem is the one least likely to be able to find the duplicate question. They don't understand their problem or its cause well enough to know the right search terms to use. Plus, veteran users of the site often have an easier time using our search system to find questions than do new users. Or they may be the same person who asked or answered that previous question, and therefore remember it—happens to me a lot.

and then is told that it's a wrong question because a different question happened to have the same answer.

Wait, what? A "wrong" question? Who said that? The closure message certainly doesn't say that. It says that this question already has an answer over here on this similar question. That's accurate.

If someone left a comment chastising the person for asking the question, that comment should be flagged as offensive and removed.

But certainly the act of closing a question as a duplicate makes no commentary on whether or not a question is valid, whether it was appropriate to ask, or any other such thing. In fact, it arguably says that it was perfectly fine, since there is already another question asked that is like it (assuming, of course, that the duplicate target wasn't closed for some other reason, but that is unusual and really only happens when the asker is the same person).

That seems really counter productive and harsh.

How, exactly, is it "harsh" or "counter-productive" to find an existing question that already has an answer to exactly what someone is asking, and provide them with that link?

In fact, I would argue that it is precisely the opposite of both of those things. It is about as productive as we could possibly be, and I see nothing harsh about it.

I don't know about you, but the reason I ask questions is to get answers. If that answer has already been provided somewhere else, and even better has already been vetted by the community of experts we have participating here, that's a win all the way around.

Of course, if you see a question that gets mistakenly closed as a duplicate, you should cast a vote (or a flag) to reopen it. That would be the case if, like you said, a more general question was closed as a duplicate of an extremely specific one.

Alternatively, if the answers to the original question are very good and are responsive to the new question being asked, you could edit the original question to be more general and treat it like a canonical question that has already received excellent, community-audited answers.

share|improve this answer
    
As was discussed when the "not a question" tag was retired, having a question closed will always be perceived as a rebuke, and it is very offputting to new users to have their question closed with a reason that isn't actually true. The current phrasing leaves a lot of room for "but that isn't the question I asked!" indignation. –  BostonJohn Aug 9 '13 at 17:33
    
I think we need to get creative here. A well titled question will be more easily searched for than a poorly titled question. A well formatted question will be more easily recognized by a later questioner as being what he really wanted to ask. The new question adds value. The old answer, being a good one, also adds value. How can we devise a scheme that captures (and maybe rewards) both the new question and the old answer? –  Walter Mitty Aug 9 '13 at 23:39
1  
@Walter We already reward people who ask questions by giving them answers. Especially the person who asked the new question, they get their answer immediately without having to wait. Isn't that enough of a reward? –  Cody Gray Aug 10 '13 at 4:40
    
@Boston Yeah, I thought that was a lot of butt-hurt over nothing, too. I cannot imagine why having a question closed as a duplicate is a rebuke. You can get offended over just about anything, that doesn't make it sensible. –  Cody Gray Aug 10 '13 at 4:40
    
@Cody, well, the new wording about "your question may have an answer here" is a big step in the right direction. Maybe we could change the close reason from "duplicate" to "previously answered". But I suppose there will always be some who remain offended. –  Walter Mitty Aug 10 '13 at 16:37
    
@Cody, people who ask bad question get rewarded with answers, too. My point is that if the old answer is a great one, and the new question is a great one, maybe there's a way to capitalize on that by leading future searchers to the new question and the old answer. Adding a reward is a secondary consideration. –  Walter Mitty Aug 10 '13 at 16:41

You must log in to answer this question.

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