On 23 Jan 2014, at 13:34, Stephen Paul King wrote:
Dear Bruno,
On Thu, Jan 23, 2014 at 4:22 AM, Bruno Marchal <marc...@ulb.ac.be>
wrote:
On 22 Jan 2014, at 23:16, Stephen Paul King wrote:
Dear Bruno,
On Tuesday, January 21, 2014 1:11:16 PM UTC-5, Bruno Marchal wrote:
On 21 Jan 2014, at 15:45, Alberto G. Corona wrote:
> It is a phisical definition of computation in the physical world,
to
> distinguish what physical phenomena are computations and what are
not.
> I don´t care about mathematical oddities.
But nobody has found such a definition. Physical computation are only
recognized as computation in machine that we can build, from subset
of
physical laws, to implement the mathematical definition.
Why not? The solution is staring us in the face. We have to
recognize that the class of Physical systems have related a class
of Representations: all of the possible measurement data of a
physical system. We can examine the measurement data and generate
simulations of the physical system in order to predict its
behavior. We call this Physics.
I don't see how this could make sense. But if it did, why don't you
use it and provide that definition of "physical computation"?
A computation is any transformation of information.
That is not the standard definition.
Information is any distinction between two things that makes a
difference to a third.
What happens when the ability to make distinctions vanishes?
We become quite retarted, I guess. I do agree with the importance of
being able to do distinction. It is basically Brouwer's first axiom
for consciousness, actually the ability to distinguish 1 from 0, if I
remember well.
Then it is a theorem that we cannot recognize something as being a
computation, even in the arithmetical reality.
Sure, but that assumes that one is dealing with an infinite set.
The set of measurable data of a physical system is not infinite.
In which theory? As long as we don't have the theory we can't say. I
assume comp, and I show the TOE does not have to axiom anything
infinite. Elementary arithmetic don't assume infinite set.
Elementary Arithmetic does not assume the Integers, implicitly? Take
the empty set, put it in a set, put the result in a set,
Or add one, again and again. This provides only finite numbers, or sets.
repeat infinitely.
Oh! Well yes, but here you assume infinity, and I said that I don't do
that.
Infinity.
... because you repeat your operation infinitely. To do this at the
base level, you need to explicitly assume an axiom of infinity (like
in the usual set theories).
We can build one and
recognize those we built, or we can bet that some process computes,
like when saying "yes" to a doctor. But there is no general means to
see if something is a computation or not, and this will depends in
part of we look at it.
This remark seems to have an interesting implication: that if I
examine some string of code that might happen to be a simulation of
a physical system, I will not be able to know which physical system
it is. We get universality of computation this way?
Computability is a notion discovered in math. It is related to the
key
discovery of Turing (also some others) of the universal (Turing)
machine.
But this universality comes with a great price. It abstracts away
time and space and all the rest of our local reality.
But we have discovered it, and it does not abstract space and time
away, it explains the persistent illusion with all possible details.
It says only that adding an axiom at that level cannot work.
Nature does not need axioms. Donald Hoffman has changed my thinking.
And I cannot use the notion of "nature". I am not sure what you mean
by nature does not need axioms. I am not sure nature needs anything,
nor even that it exists at the ontological level.
You can defend naturalism, or physicalism, and you have the right to
believe in a primitive physical universe. I am agnostic, and I have
to
be, if only because we have not yet decided between Plato and
Aristotle. We are very ignorant, notably on the mind-body question.
Umm, your agnosticism does not seem very strong. You defend AR very
strongly.
No. I debunk invalid argument against it, with some vigor, perhaps.
And yes, I do tend to believe that 17 is prime.
I get that, you defend viridity. Nature does not. Nature evolves.
In the Aristotelian theology. My point is only that this is not
compatible with comp, unless you add some magic.
I have offered you a sketch of a solution to the mind-body problem
and you vigorously attack it with demands for formalism that I
cannot write.
Only because you are using your informal and unclear ideas to
criticize the UDA's consequence.
Umm, no. I criticize its assumptions: e.g. That numbers can exist
independent of that which they reference.
If you mean that numbers -> the existence of the moon, and of the many
things which can be observed, then I agree. But the numbers are
simpler conceptually than those things referenced. So I prefer to
explain the moon by that dependence on numbers, instead of explaining
the numbers by using the moon.
What if both Plato and Aristotle are wrong?
What if you are wrong?
I am wrong. I try to correct the errors.
Good. It is the same for Plato and Aristotle.
I do not defend computationalism. I just show that IF we assume it,
then we get a constructive and testable platonic theology, which
explains physics. And I have done a piece of the derivation and
tested
it.
It does not take much to show examples of your defend, Bruno. You
are lying to yourself in claiming "I do not defend
computationalism." You will not consider any alternative.
I thought you defend computationalism also.
My case is different. I am agnostic on computationalism. But I study
its consequences. it is my job.
And, actually, I don't see any other way to even just conceive an
alternative.
That is a problem: You cannot imagine an alternative.
I just said that I can.
I said that I can conceive an alternative to comp, but only by
studying comp.
It is common that people *pretend* to defend a non-comp theory, but
they have not study comp, and actually build their attack of comp,
almost exactly following the comp first person notion provided by the
Theatetus. There is a sense in which the third hypostase (alias the
first person, S4Grz, Bp & p, ...) can be said to have a non-comp
discourse.
Your mind is closed. :_(
If you are right on metaphysical naturalism, with a real ontological
universe, then comp is wrong. That is all what I say.
Pfft, that is a false dichotomy.
Then UDA is flawed.
It is not necessary to assume ontological primitives that have some
set of properties to the exclusion of others.
Then your ontology is amorphous. Nothing can emerge from it, without
magic.
Magic is when Numbers can exist and have nothing to represent.
So you have no model for RA or PA?
I think you go to much on the meta-level.
Are you telling me that you disagree with 2+2=4?
Also, how could numbers even appear from an amorphous existence in
which nothing can be distinguished?
You are losing me.
You hold onto this dichotomy because it is your tool to defend AR.
I need indeed that 2+2=4.
>
> Computation in this sense is a manifestation of teleological
entities
> capable of maintaining his internal structure.
I can accept this as a putative truth about a notion of physical
computation, but this has not yet been defined.
Why do we need a well founded definition?
We don't.
I offer a non-well founded definition: Computation is any
transformation of Information. Information does not need to be of
physical systems; it can be of representational systems: like you
favored Sigmas and PA.
You can't change the definition. Create a new concept if you want,
but computation, or the weaker notion of computability that I need,
is well defined by Church thesis.
"reducing entropy" was
a good try, less wrong than "quantum computation" (despite here
Turing
universality is verifiable), but it does not work as nature can
compute without dissipating energy (indeed quantum computers requite
that).
Where do you get that rubbish idea?
If a quantum computer dissipates energy, the entanglement will
propagate from the environment, and the quantum information will be
lost. It has been shown (by Landauer and zurel, that only erasing
information needs energy, and logicians knows since some work by Hao
Wang, in the 1950, (I think) that universal computability can be
obtained with machine which never erase memory.
(You are Insulting. I take it that you have no argument).
Wrong! You are assuming an a priori existing infinite resource:
memory.
No. in UDA's step 0, I assume that we can survive with a body-machine,
and Church thesis.
At that stage we are agnostic on ontological infinities.
Of course we have to make sense of 0, 1, 2, 3, ...
In AUDA, we assume no more than
0 ≠ s(x)
s(x) = s(y) -> x = y
x+0 = x
x+s(y) = s(x+y)
x*0=0
x*s(y)=(x*y)+x
Then we add only definitions, in that language, using only those laws.
That is we do not assume infinity.
Comp appears to be finitist. Infinity appears in the mind of the
numbers.
We can believe in 0, and in 1, and in 2, and in 3, etc, without ever
believing in {0, 1, 2, ...}.
Quantum computation has been proven to require resources if it is
to be evaluated.
Locally. because you need to cut. But read and paste does not
require it.
Only if there is infinite memory available.
The result of Landauer-Zurek is valid in a finite universe.
Sure, the evolution of the phase is Unitary, but this holds for QM
systems in isolation. The only real example of such is the Universe
itself.
Which would be enough.
Good!
We get the Wheeler-Dewitt equation with its vanishing of time.
This go in the comp direction, although a lot of work remains to
have a clearer view on this.
I use the isomorphism between the unitary evolution of the
wavefunction and a computation.
Which isomorphism?
If you define the unitary evolution on some digital lattice, you get
an example of computation. But that does not help me to see your
"isomorphism".
> Math do not compute.
That does not make a lot of sense.
Math performs no actions on its own.
OK. Math is not even something that we can defined in math.
No self-reference?
No 3p self-references, because you need a body or a Gödel number
(name, description, etc.) and math has none, like already arithmetical
truth has none.
1p self-reference? I don't know. We can certainly not get it by the
Theaetetus' method, as math has no name such that we can define a
believer associated to it. "math" is similar to "god" here.
> Moreover it is an
> operational definition closer to everyday reality and includes all
> that is traditionally called computer science and biology (and
> sociology) within a wider physical framework.
May be. You did not provide a definition of physical computation. Nor
of "physical", which might help a skeptic like me. The only one you
gave was "reducing entropy". But it does not work. It might work for
life perhaps. It is certainly an interesting idea. But it is not
"computation". You can't change definition at will, or we are talking
about different things. The mathematical notion of computation is NOT
controversial. The physical notion of computation is not even
existing, and most attempts are controversial.
The existence of my desktop computer is obvious to me....
OK. But that "obviousness" is the mystery we can explain in the comp
theory. "obvious" is 1p, and treated in the "& p" hypostases.
Umm, OK.
Umm, OK.
Bruno
http://iridia.ulb.ac.be/~marchal/
--
You received this message because you are subscribed to the Google Groups
"Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to everything-list+unsubscr...@googlegroups.com.
To post to this group, send email to everything-list@googlegroups.com.
Visit this group at http://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/groups/opt_out.
