I have a question I am curious about: call it $Q$.

There exists a posted question $P$ on MO which asks my question $Q$. More precisely, $P$ states a series of (slightly vague) questions, one of which is very clearly $Q$.

The poser of $P$ has accepted an answer to $P$. However, neither this answer nor the other answers/comments to $P$ point towards an answer to $Q$.

Should I post my question $Q$ (being more explicit than $P$ in what I want for an answer, in hopes that this will stimulate some response), thus taking the risk that it will be unceremoniously squashed as a "duplicate" (and it certainly would be a duplicate)?

Should I edit $P$ to indicate that some part of it really hasn't been answered (though it seems rude to edit someone else's question this way, and it would still have an accepted answer)?

Should I do something else?

Should I just give up?

• Interestingly, I was just about to post a similar question to gather some community input and ultimately make this a faq. (I wonder if we were motivated by the same thing but that ultimately doesn't matter.) In view of this interesting coincidence and the fact that we generally prefer when the community brings up an issue, I will not post a competing question to yours but I will feature it so it can gather more answers. – François G. Dorais Aug 22 '14 at 22:24
• Post Q with a link to P and an explanation of how Q differs. – Gerry Myerson Aug 22 '14 at 23:22
• I suppose I could post Q with a link to P and an explanation of how Q doesn't differ. P really does contain Q. – Charles Rezk Aug 25 '14 at 22:52
• In theory you can put a bounty on P with a custom comment that explains you want an answer to Q. I don't know how well it works in practice. – Kaveh Aug 26 '14 at 1:41

I recommend posting your version $Q$ of the question, linking to $P$, and stating explicitly why the accepted answer is not what you want.
I don't like the scenario of editing question $P$ because I expect that people are less likely to answer a question that already has an accepted answer (and it feels like it undermines the original asker of $P$).
• I like this answer, but mind your $P$s and $Q$s! – Peter Dukes Aug 25 '14 at 9:31