On 3/7/2012 8:18 AM, Bruno Marchal wrote:
On 06 Mar 2012, at 19:43, Stephen P. King wrote:
On 3/6/2012 12:52 PM, Bruno Marchal wrote:
On 06 Mar 2012, at 17:53, meekerdb wrote:
On 3/6/2012 5:54 AM, Bruno Marchal wrote:
That specific retrodiction came from Bruno's hypothesis which is
that universes are generated by computation. What is computable
is much less than all mathematics.
This is not my hypothesis. It might be Fredkin or Schmidhuber
hypothesis, but not mine. My hypothesis is the hypothesis that "I
am a machine", which is ambiguous, so I put it in the form of "yes
doctor", which means that there exist a level such that my
consciousness remains unchanged for a digital functional
substitution done at that level.
And then the reasoning shows that the physical universe(s), are
not generated by any computation. Computations generated my
consciousness, and the physical universe is what my consciousness
can predict from the mixing of determinacy and 1-indterminacy in
the UD* (or sigma_1 part of arithmetic).
If I had written universes are indirectly generated by computation,
would that have reflected your view?
Better.
But the presence of the word "generated" might still lead to
confusion in this setting. Universe(s) are only observed, It is, or
they are the 'natural solution' of the comp diophantine measure
problem, which bear on the first person.
The only catch I see is that you wrote "can predict" instead of
"must predict". Are you allowing for some agency here? m
I allow for agency, but not at that level. Indeed Matter, but matter
only, is what the mind cannot act on. But the mind can act on the
mind, and agency emerges at higher levels.
Dear Bruno,
Why does it seem that you are tacitly accepting the definition of
matter as a "substance" as Descartes did in his substance dualism?
[BM]
I precisely don't do that. That's when I use the word "primitive
matter" for the aristotelian conception of matter, which is more
primary than substantial, but is still primary.
Dear Bruno,
I am trying to be consistent and agree with your explanations but
it is difficult. It is not your fault, our natural languages are biased
inherently toward certain modalities of thinking to the exclusion of
others. I was commenting on your wording, semantics.
If matter is an appearance (and not a substance), does this not allow
a form of "mind acting on matter"?
[BM]
In a large sense of that expression.
[SPK]
OK, then does this not contradict what you wrote: " Indeed Matter,
but matter only, is what the mind cannot act on." I am trying to
understand what you where thinking... I think of matter in terms of its
best representation "that whose behavior is best computationally
emulated only by itself" - following S. Wolfram's reasoning - it has a
fixed point property in this way, but it is not the same fixed point as
that of Kleene, it is the fixed point of Brouwer. It is "topological",
not "logical". The relation between them is the main feature or 'kernel"
of the process dual aspect monism that I advocate.
One only need to consider that the selection process whereby the
"next" state in time of a configuration of matter is done by a
computation.
[BM]
This does not really work. matter is a question of observable, by
machine, and the way you talk leads to the digital physics confusion,
with the idea that matter is generated by programs, when matter is
seen by programs, due to the first person indeterminacy, which bears
on infinities of computations, not just one. They might be a winner
program, but that's an open problem in the comp theory.
[SPK]
I am accepting as true the conjecture that there is no "winner
program" in any kind of global sense, there are only local optimal
winners. In this way I do not suffer from the measure problem. The local
optimal winner idea is the same as a "Strategy
<http://en.wikipedia.org/wiki/Strategy_%28game_theory%29>" that tends to
an equilibrium <http://en.wikipedia.org/wiki/Nash_equilibrium>. My
reasoning follows the same reasoning of what occurs in the question of
whether hypergames are finite or not.
A real example of this idea is implemented in the generation of
MMORPG games
<http://en.wikipedia.org/wiki/Massively_multiplayer_online_role-playing_game#System_architecture>
that are very popular. Consider the Bostrom-like
<http://www.simulation-argument.com/simulation.html> question: Since
we cannot prove that our physical reality is not a MMORPG virtual
world, should we not bet that it actually is?
[BM]
?
[SPK]
Have you seen any virtual reality generating programs and studied
how they deal with concurrency problems? Do you understand the
concurrency problem? It is basically that computations cannot
effectively solve resource allocation problems. You might be blind to
this because of your Platonist interpretation of computation and
mathematics in general... :-(
Comp precisely entails that we are in infinities of "video games". So
we can test if we are at the level zero, or if we are simulated, just
by comparing the physics then being infinity of games, which is unique
and well defined (the Z and X logics, and their higher order
extension) with what we observe.
[SPK]
This is inherently difficult because we can only access finite
computational resources to do that test in the physical world and the
test requires infinite repetition to yield non-trivial results. This is
the measure problem all over again! Do you see how the test by
falsification is almost impossible and thus your thesis that COMP is
falsifiable is very easy to argue against with weak arguments? I believe
that COMP is correct but that it is incomplete, not as a theory per se
but in its interpretation. Incompleteness is an inherent property of
non-trivial finite theories. We also have to account for the appearance
of interactions between the "stuff" of physics! The so-called
psycho-physical parallelism
<http://en.wikipedia.org/wiki/Psychophysical_parallelism>.
How do we represent interactions between the games? I conjecture
that physics is the interaction between games and all interactions occur
as bisimulations between them. (Each game is associated with an infinite
number of computations that can implement them as you point out and the
"players of the games" and the games themselves are interchangeable.)
One test for this question is to consider the upper bounds on the
ability to detect differences in features at smaller and smaller
scales. If, for example, space-time is "granular" then this would
almost certainly prove that our physical world is isomorphic to a
MMORPG.
[BM]
The contrary. Comp a priori makes matter into a continuum. You
confuse, like many, comp and digital physics.
[SPK]
I was considering a test case in my example above and did not state
this explicitly. I agree with you but was trying to demonstrate an idea
with an obviously false example.
I agree that COMP makes matter a continuum, but only in the case of
the sum over many disjoint classes of games, similar to the concept of a
multiverse in Everett and Dewitt's interpretation of QM. This is
faithfully represented by considering orthocomplete lattices as plenums
of Boolean algebras. But we have to be cautious in thinking of that idea
because there does not exist an a priori order (or pre-order
<http://en.wikipedia.org/wiki/Preorder> as the truth values are not
limited to [0,1}) on the games (we see this explicitly in the case of
hypergames
<http://weblog.raganwald.com/2007/04/repost-hypergame-and-infinity.html>
and Chu_k spaces <http://en.wikipedia.org/wiki/Chu_space>).
Within each virtual reality game there is a discrete pixelation
<http://whatis.techtarget.com/definition/0,,sid9_gci844539,00.html>
which is the substitution level for the generated content of the game.
This is what generates the "appearance" of substance. Digital physics
does not take the relativity of this into account as it tacitly assumes
a lowest upper bound on the computational resources of the "physics",
i.e. it only considers one "physics" that is implemented digitally, we
see this in Zuse
<http://www.americanscientist.org/bookshelf/pub/the-computational-universe>
, Schmidhuber
<http://www.americanscientist.org/authors/detail/juumlrgen-schmidhuber>
and Lloyd's work. COMP assumes the relativity and accounts for it in
terms of the substitution level in Yes Doctor, but suffers from problems
induced by "classical physics" thinking. (My complaints about
teleportation are a reference to this.)
This idea would be compatible with COMP if we can identify the
"players of the MMORPG" with the individual Löbian machines.
Given that some very resent observations of ultra-high energy
gamma photons indicate that space-time is not granular, we need a
more sophisticated theory to get the idea to work.
Not at all. Comp implies high plausibility of the existence of a
physical continuum, given that physics becomes an infinite sum of
infinite computations, including infinite dovetailing on infinities of
fields, including the reals. You are not yet taking into account the
role of the first person indeterminacy in the translation of the comp
body problem into a measure problem on the whole UD*, I think.
[SPK]
Maybe we have completely different ideas of what "physics" is. For
me, "physics" is the content and dynamics of a "common world of
experience" that is invariant with respect to transformations (copy and
paste operations in and between observers!) of 1p plural "content", aka
"diffeomorphisms <http://mathworld.wolfram.com/Diffeomorphism.html>". It
is the "sharable" content in the sense that all observers that believe
that they "communicate with each other" (without contradictions!) and
their belief is true (in the Bp&p sense) within their shared content.
But this only is considering the dynamics, there is also the "stuff"
that undergoes these dynamics and the appearences of such must be
accounted for. I conjecture (with Vaughan Pratt) that the "stuff"
(particles, atoms, electrons, photons, etc.) is faithfully representable
as topological spaces (not just as number theoretical relations) and
thus the relation between logics and topologies is the same relation as
that between minds and bodies. So yes, the mind-body problem does reduce
to a body problem in COMP. Pratt points this out in his papers when he
mentioned that interactions between minds and bodies is trivial, but
interactions between minds (or bodies) is not. This is the concurrency
problem (and the measure problem!) that I keep mentioning.
I see "sharing" not as an a priori relation, like set intersection,
only but also as the collection of equivalences between observers -
which I am considering in terms of games as their 1p content - is it
more like an equivalence class as a Category but with natural
transformations <http://en.wikipedia.org/wiki/Natural_transformation> in
addition to endomorphisms. There is a version of this idea in the study
of "quantum games" where it has been shown that entanglement generates
behavior that, in some limit, is identical to classical "substance
exchange" models of interaction without any actual "substance exchange".
A similar notion is found in Leibniz' notion of monads but an error in
reasoning prevented any progress there.
Consideration of this kind of idea is important if we are to
finally disabuse ourselves of the Aristotelian notion of substance.
No. The reason why "my consciousness" can predict, as opposed to
"must predict", is the first person indeterminacy. It is the fact
that I cannot know which machine I am, nor which computations
executes the relevant states.
We can have partial information set, like, assuming bla-bla-bla, if
I am duplicate in {W, M}, I will feel to be in M or in W. That is
disjuncts. But by UDA-(step 8 included), I have to say at each
instant I will be in u1, u2, u3, u4, ... that is the infinite
sequence of programs generating my current state. They all compete
in the measure, and "we" can only see the result of that from
inside. Here the 1p and its invariance for the delays explains that
such "results" never appear in the UD, but is on the border of UD*.
Does not first person indeterminacy also occur in any kind of
displacement of relative position, no matter how small that
displacement might be? But we have to consider more than one kind of
change. We have to consider relative changes for all possible
observables such that thecanonical conjugate rule
<http://en.wikipedia.org/wiki/Canonical_conjugate> is preserved.
[BM]
We don't have yet any notion of position, so your problem is not yet
formalizable in the comp frame. It is premature.
[SPK]
Yes, it may be premature, but conjecture we must or the open
problems will never be solved. I wish you would discuss with me the
Tennebaum issue that I have mentioned previously. It is part of the
reasoning of my conjecture. My main difficulty is that my thinking on
this is not in a verbal or symbolic format and so my ability to
coherently communicate it is hobbled. It is more a "picture in my head"
that I am struggling to communicate...
Onward!
Stephen
--
You received this message because you are subscribed to the Google Groups
"Everything List" group.
To post to this group, send email to everything-list@googlegroups.com.
To unsubscribe from this group, send email to
everything-list+unsubscr...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/everything-list?hl=en.
