homesafe on Nostr: I don't think I'm behaving unusually irrationally wrt my own (admittedly extremely ...

I don't think I'm behaving unusually irrationally wrt my own (admittedly extremely high) uncertainty

This is not a precise policy that I follow meticulously, but my general mode of conducting myself is to run with my best guess, including my best guesses about how likely it is that I've somewhere gone completely off track in a way that's gonna end up causing me to smash my face into an unexpected obstacle

Like, the way you become less confused is by exposing your confusion to trials where it makes contact with reality

If you want to figure out exactly which string of bits was used as input to an algorithm like the generation of a IEEE standard string - you're gonna need to go looking for more evidence that's causally entangled with the contents of that input string

And the way you do that is by having a space of hypotheses about the exact string of bits, and going and looking for observations which occur with different probabilities given the exact strings in that hypothesis space

So the correct way to deliberately choose such that your belief-state is updated necessarily means that your actions are dependent on which hypotheses your considering

Which is what I've been attempting - at no point in this conversation did you say explicitly "I am conversant in epistemology to a degree where I can evaluate your arguments" - yet I am attempting arguments based on formal epistemology anyway because one of the "hypotheses" ( / ways I'm imagining you) is that I'm talking to a person who would find these arguments legible enough to think about productively

And sure you could say that was an *easy* type of guess if it were true - or that I was foolishly running out ahead of myself if it was false - but it'd take substantially more deliberate effort on my part to do my very best not to single out any hypothesis if I didn't have enough evidence to narrow down the space of possibilities to where that hypothesis is actually distinguished enough to be worth effort considering

You're setting a bar that's higher than any human can actually achieve, and even people who do such a good job that they happen to manage to narrow down a space of hypotheses to just 2 will have to run a test between them, and if the test looks like engaging with one of them as if it were true then 50% of those worlds could have someone saying "woah hold on there, you couldn't possibly know that for certain, that's a space of around a hundred million possibilities and you only had 27 bits of evidence!"

So like, roughly, I just make guesses about who I'm interacting with and what thoughts they're having which cause them to say the things they do - and then I engage with those thoughts and see if I'm making good guesses or not

And yeah, I do it in a messy way, I don't even use a notepad and 30 minutes of clock time to review all the stuff I know - not unless I feel forced to be *really* careful

I've just tried to train my intuitions via situations where I can see roughly the correct amount & location of uncertainty I ought to be feeling - and then in stuff like social interactions, I just hope that my intuitions and my habits in translating those intuitions into actions don't end up being so disasterous that something unrecoverable happens

& It doesn't seem like this was unrecoverable - I may possibly be annoying you a little bit by being pretty blatant about assuming things about you for the purpose of choosing what to say - but I'm not seeing signals that we're about to become mortal enemies or something 😅