Every once in a while a clever person will be so convinced of their cleverness that they can miss a point artfully. This isn't really all that bad, and is probably better than being wishy-washy – except sometimes they'll start delivering personal insults to people who aren't totally on board with them. For example:

People disagreeing with Robin Hanson and Nick Bostrom must be in denial.

I suppose it's worth clarifying the first multi-meaning of the header: that Robin Hanson will draw pretty specific conclusions from polls he puts on Twitter, as if the people who are willing to wade through the flood of questions (many only slightly different from each other) are thinking things through clearly. If you want to actually engage in an analysis, polls are a terrible way to do that – just write your thoughts down. It's like he's wanting people to shut up and just characterize their ideas instead of sharing them. Which is only useful if you don't want the best idea, but instead information about what people are thinking. I don't see the point of that when you're getting the results seen above. Isn't this type of question far outside the range of applicability for any "wisdom of the crowds" style analysis?

And the weird thing is, he doesn't seem to want either wisdom-of-the-crowds or a good argument. Read his full thoughts. He just does a weird drive-by on people who think Bostrom's argument is flawed – but that will be addressed later.

The second multi-meaning is just how exhausting it must be to communicate with someone who genuinely doesn't want to engage with you. He doesn't solicit specific suggestions about what might be flawed in Bostrom's possibility-partitioning argument, or even reasoning about why we might expect a flaw to exist (because many things can trip heuristics for being wrong even if discovering the exact nature of the problem is extremely difficult). Only, 'Here's a poll. Oh, you disagree? You're an idiot.'

What a chore. Is there some amazing reverence I'm supposed to have for someone who casually dismisses people like that?

The third meaning, and the one that prompted this, is how totally certain Hanson is that he's exhaustively explored the space of possibilities. Go look at the content of the poll, and read the argument from Bostrom he's talking about. As I mentioned in the links post, while talking about regular polyhedra, it's really easy to convince yourself you've completed an exhaustive search, and orders of magnitude harder to have actually completed it.

This particular example dives head-first into things where even a logical formalism isn't enough to ensure it's true (as axioms can sneak in a lot of plausible but slightly flawed implications) – it's essentially the Boltzmann Brain of software. Our analysis of the probabilities here should be exceptionally careful.

Continuing Bostrom's argument, for instance: as simulations would be turned off once they thermalized (and are no longer interesting), and the real world can't be turned off, so the root reality dominates universe-existence time. We should expect we are neither a root civilization nor a simulation, but a Boltzmann brain, with probability 1. But that's very, very dumb. And yet it uses identical logic! [There exists a more careful formulation of this, considering both how simulations require negentropy, and simulations that might occur in Boltzmann-civilizations, but they don't change anything important.] Bostrom is smart, and I'm sure his formulation was just a jumping-off point. If you can't rule out absurd things, you aren't prepared to be exposed to them. They represent a distraction and danger akin to being radicalized by QAnon. Hanson fell into the trap of credulously reading a thought experiment and deciding that the apparent trilemma should be accepted totally at face value.

Which, again, isn't that bad, but then he's rude.

Perhaps I am in denial, Robin Hanson. I deny the intellectual value in looking down on people you won't even listen to.