> Convincing him otherwise is difficult. He admits that it's useful, but
> he thinks that most of the time it's just a pain in the ass that gets in
> his way.
He's right. The same can be said of seatbelts, too. The overwhelming
majority of the time, they're just in the way. After all, only a very
small percentage of one's driving time is spent in a collision.
But, I suppose, if you find yourself experiencing one of those crisis
moments, I'd wager you prefer wearing the seatbelts than not.
(Note: Rabidly anti-seatbelt nuts may replace this analogy with
parachutes for fighter pilots.)
His problem is that he's merely comparing the likelihood of an adverse
occurrence, but not taking into account the severity of the outcome.
Baby decision theory would suggest that we multiply the utility by the
Jesse Hughes "I AM serious about this being a short route to a Ph.d
for some of you, but just remember, I'm the guy who proved Fermat's
Last Theorem in just a bit over 6 years [...] My standards are kind
of high." --James Harris, founding a new mathematical school