Piggy backing off of "How important is a word origin?", what level of an etymology question is safe to close as general reference?
I think very few etymology-related questions would be safe to close as general reference.
Most online or printed "general reference" works dealing with etymology or word origins have existed up to this point without the benefit of Google Books. I won't argue that this tool is the be all and end all of word origins and such, but it is a very powerful tool that is just beginning to be tapped. In the process of answering some basic-looking questions on this site, I've turned up several antedatings and bits of new information on words and phrases that have in turn been used to update entries at general reference sites like Etymonline and The Phrase Finder.
As Doug Harper writes in the introduction to his Online Etymology Dictionary:
The recent availability of newspaper and magazine archives in computer files, and the flood of material presented in the searchable Google books project, opens a vast field for careful research [...] The bulk of the 19th and 20th century dates in this work have been found, or confirmed, in such sources.
The availability of Internet newsgroups archives -- again, used judiciously -- is a convenient way to find rough early dates for the most contemporary words.
I've been reading the discussions about the value of these types of questions and about the danger of having too many basic questions on Q&A sites like this. I can see this being an issue with other types of questions, but I strongly believe that in answering questions encompassing "first usage" and "historical use in print," our work here will continue to shape, or even become, "general reference."