But on the other hand you’d still be less likely to be involved by accidents caused by others—but such a rule is not Categorical Imperative-izable.
Aren’t there stable rules which are perfectly Categorical Imperative-compatible? Thinking in UDT sort of terms, perhaps the rule would be ‘flip a coin to decide whether you start driving on the first or second day’. If everyone did that, half the population would drive day 1 and half day 2, which seems superior.
Well… by such cases I kind-of meant “several-player games”. (Then there is the absent-minded driver problem… maybe if you count the driver’s self before and after the first intersection as different players… but then that becomes a variable-number-of-players game. Whatever. I guess I’ve just internalized http://lesswrong.com/lw/vp/worse_than_random/ way too much.)
Aren’t there stable rules which are perfectly Categorical Imperative-compatible? Thinking in UDT sort of terms, perhaps the rule would be ‘flip a coin to decide whether you start driving on the first or second day’. If everyone did that, half the population would drive day 1 and half day 2, which seems superior.
Huh yeah, it would. I had forgotten that in such cases good strategies can be non-deterministic.
Well, it’s not just in ‘such cases’ but in tons of games there are mixed strategies and even mixed strategies which are the Nash equilibrium.
Well… by such cases I kind-of meant “several-player games”. (Then there is the absent-minded driver problem… maybe if you count the driver’s self before and after the first intersection as different players… but then that becomes a variable-number-of-players game. Whatever. I guess I’ve just internalized http://lesswrong.com/lw/vp/worse_than_random/ way too much.)