There was a recent comment thread on this question where one of the moderators claimed that answers with overlap (note: not even exact duplicates, just "overlap") makes the question itself a duplicate.
I fundamentally disagree with this interpretation, and the few related posts I was able to find support this position. Consider the following related meta posts:
If we (as a community) take the position that any post that generates a vaguely related answer makes its parent question an exact duplicate, then shouldn't we start closing any question that offers Hash#new or
sprintf "%0.2f" as answers, since those answers clearly have overlap, even if the framing of the question or the underlying objectives are different? That's the end result of such a policy, if carried to its logical extreme.
If there is any value in searching questions at all (as opposed to just answers), then it's in questions that re-frame the issue, offer new context, or identify different objectives to which a given answer pattern may apply.
Don't Get Sidetracked
This question is not really about the referenced question itself. The question is about community policy surrounding "exact duplicates" that aren't necessarily duplicates.
Policy Question, Restated
To restate, shouldn't related questions be left open? If not, please identify a metric for when related answers should trigger a vote-to-close on a non-duplicate question.