Stephen P. King (stephenk1@home.com)
Sat, 03 Apr 1999 18:55:26 -0500
Dear Ben,
I apologize for being a bit abrupt. I just get impatient. :)
Ben Goertzel wrote:
> >[SPK]
> > I can see your point about Pratt, but in the sum of his papers, an
> >concept emerges that is paradigmic. The idea of duality is not just a
> >clarification, it is a modeling tool. Ben's thoughts unfortunately do
> >not go far enough, as of what I have read. I am not interested
> >personally in the particular details of consciousness, e.g. perception,
> >cognition, etc. at this moment. We need to merely show that
> >consciousness is derivable from our model and let others work out that
> >detail.
>
> I suspect that the details of consciousness and perception and cognition
> ARE relevant to physics, in ways that we have not yet understood.
I do completely agree with this comment! I believe that it is because
of this relevance that the work of people such as Pratt, Svozil, Wegner
and Calude is very important.
> Specifically, I suspect that the same "archetypal algebras" appear in both
> psychology and physics,
> and that this fact is not unconnected with the idea that the objective world
> is the "sum" of every entity's subjective perceptions of it.
I would go further. :) I would say that for every possible weighted sum
of a finite set of entities's subjective perceptions (or possible
mutually consistent measurements) there exists an "objective world." The
basic fact that must be understood is no single unique sum of subjective
perceptions that can bisimulate a unique "objective" universe for all.
Bisimulational equivalence is were the computational aspects come into
play. Penrose's argument that Turing Machines are insufficient to model
conscious awareness is correct, but he did not go far enough in his
train of thought. TM's fail because they are bounded by initiality, the
same initiality (and/or finality) that is assumed in Classical physics.
With the understanding of the implications of QM's irreducible
uncertainty, we can no longer operate within the naive realism of a
"single unique and finite universe" for all possible observers. Please
read http://www.cs.brown.edu/~pw/papers/coal.pdf
The Universe is NOT _______?, with respect to any finite measure, only
in the infinite limit of increasingly accurate (or convex covers ?) sets
of possible observations can we get ______?. I do not know the proper
word to describe this property; it is defined as "binary complete" and
at universal equilibrium and Hausdorff(?), among other properties, that
relate to its Chiatin's regularity. (Robert's latest posts are
relevant!) Suffice it to say that no finite observer will ever
completely model the Universe because that would require an Eternity to
compute in terms of the model's Kolmogorov complexity. The only hope is
a never ending chain of inclusive model that model It with increasing
accuracy. This is the basic idea, IMHO, of Hitoshi's derivation of the
uncertainty principle with in LS theory.
Needless to say, we have to understand that in order to assume that
observers can have a consciousness that lasts longer than a Planck time,
we have to figure out how events are strung like beads on their
individual time [world] lines. There enters a difficulty when we try to
deal with more that two events at a time. It is well known that the
scheduling of tasks in a distributed computational system, which I am
proposing a cosmos is, more often than not we will have conflicts in the
ordering. The infamous Catch 22 is an example. Pratt's Chu space
formalism plugs right into our thinking of consciousness to solve the
perennial mind-body problem or object-subject dichotomy.
http://boole.stanford.edu/chuguide.html (the site is down right now,
Vaughan must be working on it... :)
Later,
Stephen
> ben
This archive was generated by hypermail 2.0b3 on Sun Oct 17 1999 - 22:31:51 JST