So, his point seems to be that because finding the truth is complicated, it's okay to bullshit your way into hypotheses because they might turn out to be true.

So, his point seems to be that because finding the truth is complicated, it's okay to bullshit your way into hypotheses because they might turn out to be true. But at what cost? The article is pretty much justifying p-hacking and selective data usage, but for every time that leads to an unexpected breakthrough discovery, there will be a thousand other times where it leads to bad science, and stands in the way of discovering the truth. Just because Galileo did it in the days without experimental standards and with very few researchers in total, doesn't mean it's necessary or desirable in today's world. The author also seems to be ignoring the fact that 99% of the time, the intention of the bullshitter is not to boldly go forth into unchartered territory in the absence of necessary data. It is to create a picture of reality convenient to themselves, with no regard to actual observation or to the other consequences of assuming the bullshit to be true.