On Thu, Sep 17, 2015 at 9:12 PM, Warren D Smith <warren.wds@gmail.com> wrote:
I was considering it for "intelligent life whose ultimate origin is on Earth" (ILWUOIOE)
Suppose you instead wanted to do a similar calculation for "intelligent life *on* *earth* whose ultimate origin is on Earth" (ILOEWUOIOE) Wouldn't the calculation be exactly the same? What justification is there for applying this calculation to one of these rather than the other? If it's equally valid for both, then we've just shown that our descendants will not outlive our descendants on earth. Andy
not "humans" (admittedly both are somewhat undefined) . If the births of intelligent lifeforms are ordered chronologically, then with probability P you are a member of the middle P fraction of all of them. That is the so-called "doomsday argument."
So for example, with 95% chance, the total number of ILWUOIOE's there will ever be, is <20 times the total number so far.
I simply am pointing out, this argument does NOT imply the last one will happen before 20T after now, where T is the timespan ILWUOIOEs have existed so far (which is perhaps T=30Kyears) -- because future ILWUOIOEs might have much longer lifespans. That incorrect argument is sometimes stated as though it were the "doomsday argument."
In any case the doomsday argument is a comparatively unpowerful statistical argument, in the sense the confidence approaches 100% in a rather slow manner, not in an exponential manner like you usually encounter in the sort of statistics that makes me feel a whole lot better. So caveat emptor. I hope that if/when ILWUOIOEs do not all bite the dust, you will demand your money back.
-- Warren D. Smith http://RangeVoting.org <-- add your endorsement (by clicking "endorse" as 1st step)
_______________________________________________ math-fun mailing list math-fun@mailman.xmission.com https://mailman.xmission.com/cgi-bin/mailman/listinfo/math-fun
-- Andy.Latto@pobox.com