But isn't that where original research comes into play? What is the purpose of quoting ken's claim about Dalton eating hobbits? If it's in an article on Dalton, in an effort to show that Dalton might actually eat hobbits, then the problem is it's a fringe theory, because Ken isn't an authoritative source. If it's in an article on Ken, to show how dumb Ken is for thinking Dalton eats hobbits, then it's original research. Don't quote Ken to make Ken look dumb, quote an expert who talks about how dumb Ken is.
[...]
It's a reliable source for the claim that the action happened, but not a reliable source for the truth of any of the allegations made during that action. And it's *definitely* not a reliable source for the allegations' *notability*.
That's the key point. Stating Ken's claim about me eating hobbits is fine from a reliability point of view, but if Ken isn't some kind of expert on either me or hobbits, then his claim isn't notable and shouldn't be included for that reason, and that reason alone. If Ken was an expert on people eating hobbits then his claim would be notable and it should be included in any article on me.
Of course, we're not actually talking about articles, so we've ended up a little off-topic. People need to stop applying Wikipedia's policies about articles to everything else - they don't apply. Perhaps the problem is our overuse of acronyms - people end up forgetting what they actually stand for. The B in BLP stands for "Biographies". The policy is about biographies, that means articles about people, articles go in the main namespace. The BLP policy does not apply to the rest of Wikipedia. Some of the reasoning behind it does, but not all.