From: Gordon Worley (redbird@mac.com)
Date: Wed Mar 07 2007 - 12:58:35 MST
On Feb 27, 2007, at 10:18 PM, Eliezer S. Yudkowsky wrote:
> http://qntm.org/responsibility
>
> I rather like this story's reasoning. Especially the fixed-point
> theorem and its exploitation. I point out, however, that you can
> also get stability with an N-cycle.
Nice, although this still leaves us with the problem of constructing
such a device and deciding whether such a device should be
constructed (assuming it hasn't already been constructed and we're
running in a simulation, in which case the conclusion is foregone and
we just have to "wait" to compute the appropriate time).
An additional issue to consider: suppose such a device is
constructed, but then as the nesting approaches infinity, the
universes converge to a universe where the device is not
constructed. In other words, is the space of universes complete? I
see no reason why they must be, but then I have a very poor
mathematical model of a universe.
-- -- -- -- -- -- -- -- -- -- -- -- -- --
Gordon Worley
e-mail: redbird@mac.com PGP: 0xBBD3B003
Web: http://homepage.mac.com/redbird/
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:57 MDT