"Close as duplicate" - what if only the answer is a duplicate?

Question

I know it's already established that duplicate questions should be closed but not deleted, so they can act as "pointers" to the one canonical answer.

However, I think there's one important aspect that is usually glossed over:
lately I've run into a number of instances where questions have been closed as duplicates even though they patently are not. Instead, those answering the question know that the answer is a duplicate of one presented, and to save them the trouble of copying that answer, they close the question as a duplicate.

At the most basic level, this is of course absurd. Many entirely different questions can have the same answer (would you close "what is 2+2" and "what is the square root of 16" as duplicates?), but of course most real-world examples aren't as clear cut.

I feel that a really good example of what I'm talking about is this question, which was closed as a duplicate of the (somewhat contentious) C++ FAQ question here.

In this case, the questions are genuinely different. One question is simply "why doesn't my code work", and the other is "when should I use typename to make my code compile". Of course the answer is "you should use typename here to make your code work", so the answer is the same, but the question isn't. From the user's point of view, it seems to send the wrong message when such a question is "closed as a duplicate".

Doesn't this come across as "I can't be bothered answering your question, but here's a link to a generic cookie-cutter answer that might answer your question. Oh, and in the future, please don't ask questions here, it's already been done"?

I think it's important to note that the questions are so different that the person asking might not realize the relevance of the question he is being directed to. This isn't just a case of two people asking the same thing with slightly different wording, but two entirely different questions, where the answer just happens to overlap heavily.

It is also my understanding that "dupe" questions are deleted after a while, which of course means that all the effort that goes into marking such questions as dupes to avoid answering them, is really just generating more work for these people.

As I see it, there are two ways to look at this problem:

• the "close as duplicate" text doesn't actually match what's happening. The first part, "This question covers exactly the same ground as earlier questions on this topic" is simply untrue. The second part, "its answers may be merged with another identical question", seems to threaten that the question may be eliminated which would be actively harmful. (Whether or not this merger ever does happen is (almost) irrelevant. What I'm concerned with is the message it sends to the person asking.

• or the question probably shouldn't be closed as a duplicate at all, regardless of how you word it. The question is unique. Its answer just happens to overlap those for another question.

So what should be done in this case?
(some of these ideas are just user policy, others would require new functionality to implemented)

• leave everything as it is, and close the questions as duplicate when the answer is the same as that to another question,
• we could leave such questions open, and just answer them again,
• the textual description on "close as duplicate" could be revised to better cover these cases (and of course, the question being closed should never be deleted then),
• a new close reason could be created instead (a "duplicate answer" to complement the existing "duplicate question" close reason, which might make more sense to the OP whose question got closed),
• or we could try to streamline and automate the process of duplicating answers so that actually answering these questions isn't such a chore. Think of it as refactoring tools applied to answers.

To elaborate on the last item, the problem with duplicate answers is usually that we have to stop writing our answer, and go search out that other question which has an answer we can link to (and then someone else provides a less informative, but quicker, answer and we don't get any rep and sit in a corner sulking for the next week).

What if we could "factor out" commonly used nuggets of information from our answers, store them in some kind of wiki-like repository (or the usual Q&A format could work too, as long as it is made easy for us to reference), so that we can reference it easily when writing answers that rely on this information?

In the example I linked to above, what I really want to do is just provide a simple answer "shell" containing a bit of context which relates the information to the OP's problem, and then for the meat of the answer, I'd just like to reference some extremely commonly used information. And I'd like to be able to do so quickly, without having to search, and without having to type out an entire URL.

What I really want is to write an answer such as this:

you need to use the typename keyword as described [here](wiki:c++-typename)

and then have the [here] transform into a link to the (currently fictive) SO-wiki article on the typename keyword in C++. Or even better, it could expand into an inline copy of that article, making it easier and more convenient for the reader of the answer, while still allowing us to reuse things we already wrote, and allowing us to centrally keep this information nugget up to date.

Thoughts? How much of a problem do you think this is? And how do you think it should be dealt with?

Answer 1

Hah, and 5 seconds after asking this, I discover the latest blog post. In particular, this seems very relevant to the discussion:

If you’re going to close a user’s question as a duplicate, it has to be a real duplicate. For example, if a user asks, “What does the IP address 128.0.1.1/24 mean?” it’s OK to close that as a duplicate of a more general question like “What do IP addresses of the form a.b.c.d/e mean?” But it’s not OK to close it as a duplicate of a twenty-seven page guide to netmasks. That’s the moral equivalent of saying “RTFM.”

Stack Overflow is not meant to be a library of reference manuals. It’s supposed to contain the same information as a library of reference manuals, in the form of millions of questions and answers. Combined with Google, that gives us the magical power of a library of reference manuals you never have to read! It’s like, you got to the library, and there’s a wizard there at the door, and you ask your question, and, instead of being told to read a book, you just got (are you sitting down?) the actual answer!

I think this rampant "close as duplicate"-ism is pulling in the wrong direction here. People are being told to go read a book instead of getting an answer. It is a slightly more polite RTFM.

Answer 2

If the correct answer is duplicated, then I think that's grounds for investigation of a possible duplicate question, not a definitive sentence of duplication.

We've had some similar points of discussion from here on Meta Stack Overflow, which in itself was cited on Gaming for "Should different questions that yield similar/duplicate answers be closed?". To quote from the Meta question,

If you ask a question similar to another question and it is likely to get the exact same answer, you have yourself a duplicate question.

Now, the key thing here is that there's two components here - getting the same answer and being similar. The presence of one is often a clue to check for the other - so if the answer to one question is the same as an existing one, that's a clue to check for if it might be a duplicate.

But there exist several cases where two questions can attract the same answer without being duplicates. It all boils down to the second point, that they have to be similar. If the differences are superficial, then it makes sense to close them. But here are examples of when things might attract the same answer but it would be rather callous to call them duplicates.

• Kaestur Hakarl's answer to the Gaming question supplies 3 examples of situations where the answer to two or more questions are the same on a basic level, yet the questions are completely different. Sometimes a particular operation works as a solution to multiple problems - that doesn't make all those problems the same. Otherwise, "A large enough explosion" would be an answer that'd create a lot of duplicates for what it can solve.

• Consider the situation where someone asks "Does variable B affect algorithm A?", and an answerer goes out of the way to make an excellent answer that describes the full of what affects algorithm A. This answers the given problem, but it also answers the question of "How does algorithm A work?", which has not yet been asked. When the latter question gets asked, it seems very awkward to close it as a duplicate of the former - who looking for general info on algorithm A would think to find their answer in a question that's focused on variable B?

Answer 3

We've already discussed this in the chat and my take on the issue is that

1. it isn't really clear whether "duplicate" means "duplicate question" or "duplicate answer",
2. I don't like having many identical answers to what's essentially the same question, and
3. I'm fine with having many different questions that are linked to the same answers.

There's a whole category of problems users run into which, although they arise from the same cause, and thus warrant the same answer, don't result in the same question, since formulating this question would require understanding the cause of the problem. The (in)famous "Why is the result of i = ++i + i++ not blah?" in the c++ tag is one of these. Once it's understood, it's easy to formulate the right question ("Isn't there a sequence point between the operands of +?), but since whoever runs into this doesn't understand it, this will be asked again and again.
I really see no point in answering these questions over and over. One important reason is that the answers come in widely varying degrees of completeness, clarity, and comprehensiveness. It's much better for the community to put its effort into creating one or two very good answers. I think it would be much better to link all such questions of one category to a (set of) very good answer(s).
Yes, often it would be good to keep the questions around, pointing to this one good answer.

So for me the problem is with questions getting deleted after they got closed as dupes. Deleting a duplicate is OK only if the question was already asked (almost) literally. Deleting a question that got closed as a dupe is not OK if the question warrants the same answer, even though whoever asked it didn't know this.