This question is spurred by a recent question on the main site: Why does some coffee "hit me" harder than other coffee?
The original (see the history) was specifically about one brand; it received several close-votes for "opinion-based" reason.
I edited the question in an attempt to make it "more objective" -- e.g., what affects coffee's stimulation? I provided an answer about the tangible aspects of caffeine, which turned out to be more of a survey, predominantly with links to other questions about factors that affect caffeine content in coffee. The question was closed, but it has now received re-open votes, so it looks like we have disagreement (at least in part, a good thing)!
Such "survey" questions and answers are not always on-topic, and are often controversial (another example is this question of mine from Stack Overflow, in which I summarized a handful of other questions). That question was closed and subsequently re-opened, meaning that it probably received several close- and open-votes.
I happen to (personally) appreciate those kind of question/answer when appropriate; I value them as an index and summary of several questions, often comparisons of (or distinguishing between) questions within a certain tag.
To the extent that this represents an example of a "survey of other questions" answer: Are such "survey" answers (and the related question) reasonable? Or should such questions be discouraged, and limit to more narrow, on-point questions about a similar topic?