I agree that canonical textbooks are superior--often because they are in the canon for a reason! But if for no other reason, it's super helpful to have everyone literally on the same page. If you say you learned Calculus and then you mention a bunch of stuff I never heard of, or don't know a bunch of stuff that I regard as Calculus, then that quirky book might (or might not) have a cute angle on the subject. But either way, it's causing too much chaos.
However, books that started off quirky sometimes become canon. And it happens from people picking it up and giving it a shot, and it works better than the old ways. There should be a path for that to happen and we shouldn't disincentivize nonstandard approaches too much.
So ... I dunno. I say maybe let these posts die a natural death of nobody answering because nobody knows what that person is talking about? Or maybe have a way of closing a question because it is "execessively nonstandard"? Don't outlaw all nonstandard books, but maybe just the ones that are beyond some sort of pale? Definitely if it's primarily circulated by someone's uncle convinced that his ancient ancestors invented Calculus thousands of years before Newton.