Entmoot principle: debate should be conducted at the slowest possible rate (including constraints arising from needing to act in time). Attempts to accelerate past that rate can generally be assumed to be bad faith.
Analogical skepticism principle: Never trust a first-principles conclusion that has not been validated by a good analogy or three, unless you’re doing pure math.
Doubt-and-standards principle: when faced with a competing objective, compromise your outcome standards relative to your objective in proportion to how doubtful you are about the correctness of your thinking.
2
5
You’re unable to view this Tweet because this account owner limits who can view their Tweets. Learn more
Not quite, that's the reductive rationalist takeaway on the analysis side. I'm talking about compromising objectives, not an abstract statistical "confidence"