Duplicates where answer is not easy to find in old question

Question

There have been a few questions recently that I voted to close as duplicates and also answered, because of the following:

I think the old question includes a valid answer to the new question, but I don't think it's trivial for the asker to identify how it answers their specific question even after we point them to the duplicate.

For example:

• I marked How do I negotiate an offer for PhD funding as a dup of Negotiating PhD funding. Roughly, the answer is the same (it can't hurt to ask if you are polite about it). But I think it might be onerous for the asker of the new question to identify the "nugget" that applies to his scenario, from among the advice that only applies to the old question.
• I marked Is IJCA a valid open access journal? as a dup of How do you judge the quality of a journal?. The answer to the former is basically a distillation of the latter; but I don't think the asker of the new question could have done that distillation himself.

How do we want to handle questions like these?

On the one hand, we don't want to spread answers to essentially the same question across duplicates.

On the other hand, we want users who come to this site and ask valid questions to feel like they got an answer. We don't want to tell them, "There's an answer, but you'll have to expend significant effort to dig through another question's answers to find the part you came for" (if they're even capable of doing so).

Also, sometimes there is an answer that is a better answer to the new question than any of the old answers; but the new answer wouldn't be a good answer to the old question.

Answer 1

I would strongly discourage posting answers to questions identified as duplicates, for the main reason that such answers will not be directly visible in the original question. For instance, if you look at How to politely negotiate PhD funding offer?, it's not straightforward that there is a duplicate of this question (How do I negotiate an offer for Ph.D. funding? is only mentionned as a linked question). Hence, two cases are possible:

• Either the content of the answer for the new question is already present in the old one, in which case this answer should not be posted;

• Or the new question is not strictly speaking a duplicate of the old one, since the best answer is somehow different, in which case it should not be identified as duplicate, but only as related questions.

As aeismail said here: "In the long run, Stack Exchange sites are not just about answering people's questions, but providing long-term curating of the answers.", and I completely agree with this view. I would think that a good curating of a topic (e.g., negotiating PhD funding) should imply to have all relevant answers at the same location. If new content is required, then as much as possible, the old one should be edited in order to improve it.

In other words, if you believe there is a slightly different point in the new question, then add it to the old question, edit the answers accordingly (or add a new answer to the old question), rather than posting an answer for the new question. I believe this is the main interest of this site, compared to traditional forums/discussion boards. As an analogy, you don't see two similar articles on Wikipedia because two authors have different views, you see a single article that includes the different views.

I don't see any problem with telling new users "There's an answer, but you'll have to expend significant effort to dig through another question's answers to find the part you came for", and I would add "Please edit the existing content in order to clarify your point, so that the next person looking for your question can benefit of your work", because that's the whole point of having a community-driven site.

---

There is a question you didn't ask, and I'm not claiming you had this in mind, but it should be observed that the current reputation mechanism basically goes against the point I just made, since posting a new answer might increase your reputation, while editing an old one does not. I'm not saying you should care about reputation, but nevertheless, it's an indicator of "good behaviour". There has been some discussion about giving more rep for editing (Reputation for editing?), but the counter-argument is that it could open the door to useless and noisy editing.