From: Eliezer S. Yudkowsky (sentience@pobox.com)
Date: Sun Jul 30 2000 - 10:37:54 MDT
"Robert J. Bradbury" wrote:
>
> > "This could lead to the manifestation of systems that would have all
> > of the capabilities of the evil "liquid metal" robot depicted in the
> > movie Terminator 2, ..."
And more. "So why didn't it form a bomb and blow us up?" "A bomb
contains chemicals, moving parts."
> Huh? If humans allow the creation of nanotech with insufficient
> error-correction-codes (ECC) in designs/programs to allow this to
> occur, then they get what they deserve.
ECC is unsatisfactory as a control mechanism. What's needed is
nanomachines that just *die* when a single bit flips, not nanomachines
that attempt to correct the error.
I also worry about prions - i.e. a deformation of a motor arm that
results in the manufacture of nanomachines with a similar deformation.
This is evolution of a sort, and it can proceed without a single
informational error in the assembly instructions.
> Blech, Phooey. All you have to have done is existed for a few million
> years longer than humans and developed slow-pokey world ships to have
> colonized the galaxy. It doesn't take nanotech.
Valid objection - since the Fermi Paradox applies to lightsails and
generation ships, it cannot be used to disprove the existence of higher
technologies.
> Furthermore, the
> "nanobot" colonization perspective indicates that he clearly does not
> understand the possible problems that nanomachinery has with the high
> radiation fluxes encountered in space.
Invalid objection. While nanosystems of the sophistication described in
_Nanosystems_ might keel over and die faced with high radiation fluxes,
ultra-sophisticated nanotech of the type produced by a human
civilization possessed of matter compilers for centuries or millennia
should be far more tolerant of radiation than biology, or possibly even
sheet metal, since sheet metal isn't self-repairing.
Besides which, nanomachinery has fewer problems with radiation than you
might think. I seem to recall certain individuals on this list telling
me that nanonukes were impossible because of the radiation; well, I met
Robert Freitas at the Spring 2000 Gathering and he said that was a load
of hooey (paraphrasing liberally, here - what he actually said is that
nanomachines could perform isotope separation on uranium without any
difficulty, and in fact would be one of the most efficient means of
doing so).
> > "... we are driven toward concluding that (a) there is no other
> > intelligent life in the galaxy or (b) non-organic nanotechnology
> > of the self-replicating micro-Babbage-robot type described by Drexler
> > is impossible. Since we know that the evolution of intelligent life
> > is posssible, but we do not know that nanotechnology is, I must consider
> > (b) the more likely alternative.
>
> Great! Using the Fermi Paradox to explain why nanotechnology is
> impossible. Now I've seen everything! Could there not be a (c)
> "Nanotech civilizations can observe almost everything, so there is
> no need to travel", or (d) "Nanotech civilizations spend their time
> trying to engineer smaller selves. They do not interact with
> pre-limts-of-physics species." etc.
No, (c) and (d) are as silly as (a) and (b). Nobody has ever given a
satisfactory explanation of the Fermi Paradox. That's why it's a
Paradox. All such a Paradox can prove is that a given picture of the
Universe is incomplete; it cannot be used to disprove specific facets of
a scenario.
--
sentience@pobox.com Eliezer S. Yudkowsky
http://singinst.org/beyond.html
This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 15:30:14 MST