From: Eliezer S. Yudkowsky (sentience@pobox.com)
Date: Thu Sep 12 2002 - 20:24:39 MDT
Ben Goertzel wrote:
>
> Mapping a person's words into a model of their future words is not that
> easy. Essentially, it requires mapping their words into a model of
> their mind. This is a very hard "inverse problem." I require more
> data in order to believe I have a reasonably good model of someone,
> than you do. Those 2 messages of GK did not give me enough data to
> form a plausibly confident model of the guy's mind.... Either you're
> way better at mind-modeling based on scanty data than I am, or you're
> just quicker to jump to conclusions in this regard...
Or I'm judging against a different criterion than you use, one with a much
narrower aperture, so that available data was sufficient to determine
(with margin for error) that the model thus mapped was outside the aperture.
PS: Cliff, the fact that Ben and I are speaking cryptically should be
enough information for you to determine what we are speaking about. Hint:
It has nothing to do with religion. (Unless I've mapped Ben incorrectly
and he *is* talking about religion.)
Crypticism can sometimes be very useful, and I won't tell you when, but
it's *always* fun.
--
Eliezer S. Yudkowsky http://intelligence.org/
Research Fellow, Singularity Institute for Artificial Intelligence
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:41 MDT