Re: Doomsday Example

Robin Hanson (
Fri, 28 Aug 1998 10:20:15 -0700

Nick Bostrom writes:
>> Consider an exponential model of the time to evolve life on a planet,
>> where the probability of life evolving t years after the planet forms
>> is proportional to exp(-t/h) within the window [0,T]. The average
>> number of planets with life is then proportinal to 1 - exp(-T/h).
>> The observation that t ~ T/2 exponentially suppresses the posterior,
>> relative to the prior, for h << T. Values of h >> T are only linearly
>> suppressed if one expects one is more likely to appear in a universe
>> that has more life. In my paper I was focusing on the exponential,
>> not the linear, suppression, but I admit both effects are there.
>Could you elaborate a little on how this linearly suppression comes

It is just the effect we have been discussing, that the fact that one is alive suggests we are in a universe with more life. The probability of a universe with N people, relative to a universe with M people, is enhanced by a factor of N/M.

Robin Hanson RWJF Health Policy Scholar, Sch. of Public Health 510-643-1884 140 Warren Hall, UC Berkeley, CA 94720-7360 FAX: 510-643-8614