[extropy-chat] Inside Vs. Outside Forecasts
wingcat at pacbell.net
Wed Oct 12 16:16:19 UTC 2005
--- "Eliezer S. Yudkowsky" <sentience at pobox.com> wrote:
> Let me rephrase: The outside view always wins over the inside view.
> Naturally, predictions sometimes fail. It's what they do. Have you
> evidence that, in the 60s, I could have done systematically better by
> accepting "inside views" of computing - stories told about how, in
> particular, computing power would grow - over outside views? It is
> enough to find one inside view that turned out to be correct. That
> could be luck. If enough people make enough different guesses one of
> them is bound to be correct.
Actually, luck is enough to invalidate "always". "Always" means 100%,
no chance of luck messing things up. So it is, in fact, enough to find
just one inside view that turned out to be correct: just one
counterexample is all that is needed to disprove a logical absolute.
I will grant that it is possible that most, maybe all, of the specific
cases where the inside view wins can be argued to be luck. I do not
have any data to prove that, in any given case where this has been
measured, the fraction of outside-view predictions that proved correct
did not exceed the fraction of inside-view predictions that proved
correct. Then again, is there data to prove the opposite case, which
you are asserting? Has outside view vs. inside view actually been
statistically measured in any given case?
More information about the extropy-chat