Ultratechnologies (was: The inevitable limitations...)

From: Eliezer S. Yudkowsky (sentience@pobox.com)
Date: Thu Nov 23 2000 - 14:04:35 MST


xgl wrote:
>
> but this sounds like another way of saying that omniscience is
> impossible ... which, on this list, is surely not in dispute.

I dispute it!

Omniscience is not known to be either impossible or possible.
My list of material ultratechnologies goes something like this:

* Nanotechnology
  + Effective material omnipotence from a hunter-gatherer perspective
  + Quantum computing
* Megascale engineering
  + Dyson spheres, ringworlds, et al.
* Femtotechnology
  + Like nano, but with baryons instead of atoms
  + Transmutation of elements, new materials ("artificial atoms")
  + Probes that can land on a neutron star
  + Refinements: Higgsium, monopolium
* Chromotechnology
  + Like nano, but with individual quarks
  + artificial baryons
  + fastest achievable purely material computers (probably)
* Spacetime engineering
  + manufacture of negative matter
  + time travel
  + violation of thermodynamics (permitted by either of above)
  + nonmaterial computing, possibly at infinite/transfinite speeds
  + wormhole FTL
  + tardis warp FTL
  + prevent/outlive Big Crunch
  + DIY Big Bang
  + Linde Scenario / Alpha Line computing
* Reality engineering / descriptor theory
  + root permissions on the Universe
  + violation of conservation laws
  + effective material omnipotence from a modern-day perspective
* Ontotechnology
  + "change the type of things that can be real"
  + First Cause engineering
  + treat the laws of physics as material, manipulable objects
  + qualia engineering, objective morality
  + violation of information-theoretical laws (?)

Genuinely flawless "self-omniscience" probably requires ontotechnology, as
it violates the information-theoretical rules that govern our
Turing-computable macroscopic physics.

Unlimited computer memory and power, in the exponential or Moser-polygon
or Knuth-arrow sense - i.e., you just stop caring about computational
resources - is probably at the "spacetime engineering" level, where you
can manufacture arbitrary, balanced amounts of negative and positive
matter and knit them into a structure where computing elements are in
arbitrarily close proximity. Your processor contains as much positive
matter as the black hole at the center of the Milky Way, but it has an
equal amount of negative matter, all in a Van Den Broeck tardis warp (*);
so it fits in your pocket, weighs three grams, and doesn't collapse into a
black hole.

(*) http://www.eurekalert.org/releases/ns-frl060899.html

The visualization I usually use when talking about the Sysop Scenario is
the nanotechnological one. Personally, I think it very likely that it
goes up to spacetime engineering, although possibly without
chromotechnology. I would be surprised to find reality engineering
possible, or anything else permitting true global instant modification of
our Universe, due to the Fermi Paradox (and Special Relativity); I would
not be surprised to find *anything* possible on a local scale, up to and
including the creation of new Universes with entirely different physical
laws; I also think it very likely that local ontotechnology is possible.

There is a very significant probability - which, IMO, can never go below
"half-and-half" - that EVERYTHING in the above list is fundamentally a
part of the "human regime", having no more relevance to the true
technologies of the future then the Victorian vision of bigger and better
steam engines.

*BUT*...

I still think you can get effectively perfect self-knowledge on a scarcely
transhuman processor - without nanotechnology, even. Another post to
follow - possibly in the next hour, possibly several days from now -
talking about the mathematics of effective perfection.

-- -- -- -- --
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 - 04:00:35 MDT