Wolfram Model (New Foundations of Mathematics and Physics)

106 views
Skip to first unread message

Philip Thrift

unread,
Aug 5, 2020, 2:17:48 PM8/5/20
to Everything List

Bruno Marchal

unread,
Aug 6, 2020, 7:36:58 AM8/6/20
to everyth...@googlegroups.com

> On 5 Aug 2020, at 20:17, Philip Thrift <cloud...@gmail.com> wrote:
>
>
> (HyPE = Hypergraph Programming Engine ?)
>
> https://www.wolframphysics.org/bulletins/2020/08/a-candidate-geometrical-formalism-for-the-foundations-of-mathematics-and-physics/
> Formal Correspondences between Homotopy Type Theory and the Wolfram Model


That is no so bad.

To be sure, the idea that metamathematics is a sort of pre-mathematics is naïve, and does not make much sense. The whole point of Gödel and the logician was the discovery that a large part of metamathematics can be done in mathematics.

Incidentally, their use of the Curry-Howard isomorphism might be more interesting than many others, but it is incompatible with the type of logics imposed by Incompleteness, where the CH isomorphism works only for the first person modes (S4Grz1, cf []p & p).

They are still missing that physics needs the full machine theology to be, not just recovered, but explained.

Bruno



>
>
>
> cf.
> https://writings.stephenwolfram.com/2020/07/a-burst-of-physics-progress-at-the-2020-wolfram-summer-school/
>
> @philipthrift
>
> --
> 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-li...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/1dff8b68-ed0f-49da-91d5-23f4e518e78bo%40googlegroups.com.

Lawrence Crowell

unread,
Aug 6, 2020, 7:54:33 AM8/6/20
to Everything List
In reading the first of these I run into the usual sense or difficulty with Wolfram of understanding how to compute or calculate things.

This does get into HoTT (homotopy type theory) which I see as a sort of quantum of homotopy or index that represents the obstruction to diffeomorphisms on paths. A hole or "horn you can't pull the reins over" that prevents any diffeomorphism that moves a curve past the hole or horn, defines a first fundamental form π_1(M) = ℤ. The HoTT is a binary set of paths that wrap around the obstruction and those which do not. In a quantum mechanical form this can be a form of quantum bit. 

The role of topology with quantum mechanics is not fully understood. An elementary particle is really a set of quantum states or numbers, and these may have topological definition. The charge, spin, etc are topological quantum numbers, and the Cheshire Cat experiments illustrate how these are in a form of entanglement. Elementary particles are really not that different from quasiparticles in condensed matter physics'

LC

Philip Thrift

unread,
Aug 6, 2020, 8:06:55 AM8/6/20
to Everything List


What is really going on here is that a language of hypergraphs  (not well specified) is what is assumed to be defined. All of fundamental physics is to be rewritten in this language, replacing the others.



By the way, when it comes to mathematics, even the setup that we have is interesting. Calculus has been built to work in ordinary continuous spaces (manifolds that locally approximate Euclidean space). But what we have here is something different: in the limit of an infinitely large hypergraph, it’s like a continuous space, but ordinary calculus doesn’t work on it (not least because it isn’t necessarily integer-dimensional). So to really talk about it well, we have to invent something that’s kind of a generalization of calculus, that’s for example capable of dealing with curvature in fractional-dimensional space. (Probably the closest current mathematics to this is what’s been coming out of the very active field of geometric group theory.)

@philipthrift

Lawrence Crowell

unread,
Aug 6, 2020, 2:04:48 PM8/6/20
to Everything List
This continuum limit might be a way that Wigner, Wigner’s friend and the cat might form tripartite entanglement. This is (|00〉 + |11〉)|W〉 → |000〉 + |111〉 where |W〉  is Wigner’s state that couples to the bipartite Bell state |00〉 + |11〉 to form the GHZ entangled state. The process is removed by a form of topological obstruction, where this coupling is some local operation that transfers “quantum phase” to the bipartite entangled state so it is now entangled with Wigner’s state. The following diagram illustrates this 

 wormhole 3-way.png

The original Bell state is the red and green, where starting from c we then unwrap this into the diagram at the left. We can then refold it again so either red or green are on the outside and blue/green or blue/red are inside.  The middle diagram can be collapsed so the colored openings squash to a point and this is a diagram for the tripartite.

This is a topological map of an entanglement into something that is cobordant. We may think of this as a map of the tripartite entanglement into a spacetime configuration of a wormhole. The bipartite entanglement is mapped to a simple wormhole and the formation of this entanglement is equivalent to the formation of a three-way wormhole. 

This three-way tubular construction can be the fundamental joints for nontrivial graphs. I can imagine well enough a 4-way entanglement where I topologically move the new colored ring, say colored yellow. Way out near blue. I then slice the diagram into two so we have two 3-way diagrams. The boundary that is produced “constructs” two states. This means we have two tripartite states of the form |000_〉 + |111_〉 and |00_0〉 + |11_1〉, here _ mean “blank” or unknown. We then have a sort of fusion rule that would produce the entanglement |0000〉 + |1111〉 + |0010〉 + |1101〉. This again requires a local operation of the LOCC variety to work.

We now put these tinker toys together to generate a large graph. The actual nodes of the graphs are not the quantum states, but more the internal branching region. A large N-graph corresponding to an N-tangle can be fused with others and indeed even closed so the quantum states are in effect hidden away. They are no longer available to experimentation. 

The more I think about this there are some interesting prospects. We might have by this a means to construct in a large N limit N → ∞ a continuum that is classical spacetime.

Philip Thrift

unread,
Aug 6, 2020, 5:17:44 PM8/6/20
to Everything List
Ultimately this is not really about "theory" (in the usual sense) at all, It is about defining a programming library: So it's really in the end a programmer's manual.

Wolfram Physics Project Functions


(an extension of the Wolfram language/ecosystem)

All of general relativity, quantum mechanics, and whatever comes next is to be written as programs in this library/language.

@philipthrift


Lawrence Crowell

unread,
Aug 7, 2020, 7:38:33 AM8/7/20
to Everything List
That might be, but a programming language that has no context with anything is not that valuable. At least it is not that valuable to me. My point is this seems to connect with concepts of spacetime as built up from large N entanglements.

LC

Philip Thrift

unread,
Aug 7, 2020, 10:19:12 AM8/7/20
to Everything List

Q. How does quantum entanglement occur in your models?

A. Two global Wolfram model states are said to be “entangled” if they share a common ancestor in the multiway evolution graph. Since spacelike-locality is not a necessary condition for branchlike-locality, it is possible for these states to be causally connected (i.e. to be connected in the multiway causal graph) even if they are not spatially local. This is the essence of quantum entanglement as it occurs, for instance, in the context of Bell’s theorem.


@philipthrift

Bruno Marchal

unread,
Aug 7, 2020, 10:39:30 AM8/7/20
to everyth...@googlegroups.com
On 6 Aug 2020, at 14:06, Philip Thrift <cloud...@gmail.com> wrote:



What is really going on here is that a language of hypergraphs  (not well specified) is what is assumed to be defined. All of fundamental physics is to be rewritten in this language, replacing the others.



By the way, when it comes to mathematics, even the setup that we have is interesting. Calculus has been built to work in ordinary continuous spaces (manifolds that locally approximate Euclidean space). But what we have here is something different: in the limit of an infinitely large hypergraph, it’s like a continuous space, but ordinary calculus doesn’t work on it (not least because it isn’t necessarily integer-dimensional). So to really talk about it well, we have to invent something that’s kind of a generalization of calculus, that’s for example capable of dealing with curvature in fractional-dimensional space. (Probably the closest current mathematics to this is what’s been coming out of the very active field of geometric group theory.)


If Mechanism is assumed, there is no choice: the laws of physics are given by the laws of self-reference, and they are invariant for the choice of the ontological theory, or first principles.

It is a way to explain the impact of incompleteness on physics: the laws of physics are “machine” or “theory” independent.

This does not mean that some choice cannot help in the derivation process, but that should be clearly made “temporarily”, and that choice must be either discarded or justifies, from the view of any universal system, run by any universal machinery. Indeed the physical reality emerges from the statistical interference based on *all* computations.

Bruno





@philipthrift

On Thursday, August 6, 2020 at 6:54:33 AM UTC-5 Lawrence Crowell wrote:
In reading the first of these I run into the usual sense or difficulty with Wolfram of understanding how to compute or calculate things.

This does get into HoTT (homotopy type theory) which I see as a sort of quantum of homotopy or index that represents the obstruction to diffeomorphisms on paths. A hole or "horn you can't pull the reins over" that prevents any diffeomorphism that moves a curve past the hole or horn, defines a first fundamental form π_1(M) = ℤ. The HoTT is a binary set of paths that wrap around the obstruction and those which do not. In a quantum mechanical form this can be a form of quantum bit. 

The role of topology with quantum mechanics is not fully understood. An elementary particle is really a set of quantum states or numbers, and these may have topological definition. The charge, spin, etc are topological quantum numbers, and the Cheshire Cat experiments illustrate how these are in a form of entanglement. Elementary particles are really not that different from quasiparticles in condensed matter physics'

LC

On Wednesday, August 5, 2020 at 1:17:48 PM UTC-5 cloud...@gmail.com wrote:

(HyPE = Hypergraph Programming Engine ?)

https://www.wolframphysics.org/bulletins/2020/08/a-candidate-geometrical-formalism-for-the-foundations-of-mathematics-and-physics/
Formal Correspondences between Homotopy Type Theory and the Wolfram Model


cf.
https://writings.stephenwolfram.com/2020/07/a-burst-of-physics-progress-at-the-2020-wolfram-summer-school/

@philipthrift





--
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-li...@googlegroups.com.

Bruno Marchal

unread,
Aug 7, 2020, 10:43:38 AM8/7/20
to everyth...@googlegroups.com
On 7 Aug 2020, at 13:38, Lawrence Crowell <goldenfield...@gmail.com> wrote:

That might be, but a programming language that has no context with anything is not that valuable.


At least, we should take the semantic of the reality on which that language is based. A language per se is not enough.





At least it is not that valuable to me. My point is this seems to connect with concepts of spacetime as built up from large N entanglements.

I agree, and the entanglement must be explained from the first person indeterminacy, singular and plural, which are imposed by incompleteness on all “creatures” living in arithmetic (or at its internal phenomenological border given by the self-reference mode available to the universal machine.

With mechanism, physics is a branch of machine biology (or psychology, or better “theology” …).

Bruno





LC

On Thursday, August 6, 2020 at 4:17:44 PM UTC-5 cloud...@gmail.com wrote:
Ultimately this is not really about "theory" (in the usual sense) at all, It is about defining a programming library: So it's really in the end a programmer's manual.

Wolfram Physics Project Functions


(an extension of the Wolfram language/ecosystem)

All of general relativity, quantum mechanics, and whatever comes next is to be written as programs in this library/language.

@philipthrift






--
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-li...@googlegroups.com.

Philip Thrift

unread,
Aug 7, 2020, 11:53:15 AM8/7/20
to Everything List
On Friday, August 7, 2020 at 9:43:38 AM UTC-5 Bruno Marchal wrote:
On 7 Aug 2020, at 13:38, Lawrence Crowell <goldenfield...@gmail.com> wrote:

That might be, but a programming language that has no context with anything is not that valuable.


At least, we should take the semantic of the reality on which that language is based. A language per se is not enough.





At least it is not that valuable to me. My point is this seems to connect with concepts of spacetime as built up from large N entanglements.

I agree, and the entanglement must be explained from the first person indeterminacy, singular and plural, which are imposed by incompleteness on all “creatures” living in arithmetic (or at its internal phenomenological border given by the self-reference mode available to the universal machine.

With mechanism, physics is a branch of machine biology (or psychology, or better “theology” …).

Bruno


But that has nothing to do with physics.. Physics only describes (in a language).

Musing about its "meaning" is for philosophers to waste their time on.

"Our best computer simulations, accurately describing everything [in physics], use only finite computer resources by treating everything as finite."
-- Max Tegmark

@philipthrift 

Lawrence Crowell

unread,
Aug 7, 2020, 2:23:38 PM8/7/20
to Everything List
On Friday, August 7, 2020 at 9:19:12 AM UTC-5 cloud...@gmail.com wrote:

Q. How does quantum entanglement occur in your models?

A. Two global Wolfram model states are said to be “entangled” if they share a common ancestor in the multiway evolution graph. Since spacelike-locality is not a necessary condition for branchlike-locality, it is possible for these states to be causally connected (i.e. to be connected in the multiway causal graph) even if they are not spatially local. This is the essence of quantum entanglement as it occurs, for instance, in the context of Bell’s theorem.


@philipthrift

I have not finished reading the main page yet. I got about half way through it. It seems two states are entangled in his system if they share a node. This is slightly different. The usual notation is that *------------* denotes two entangled states. Now, these are equivalent to a wormhole or can be transformed into that picture, where there is a center. That might serve as the node in Wolfram's picture. This might also connect with tensor network ideas and the Ryu-Takayangi theorem. The connection to locality and causality, which is tricky, is that in a pure entanglement there is no scale. It is nonlocal and everywhere. The spacetime dual of that would be conformal invariance or conformal blocks of states. 

LC

Lawrence Crowell

unread,
Aug 7, 2020, 2:35:27 PM8/7/20
to Everything List
Context is all if you are doing science, for in science we study objects and events. If your interest is in doing pure mathematics or computer science that is fine, but it in of itself does not give physics. Feynman made some note of this. I found this little science fiction clip interesting along these lines. It is about a dormant computer system activating an attack sequence in a war that is long over. Note who in a sense "won the war." The machines activate algorithms with no context to reality.


LC

On Friday, August 7, 2020 at 9:43:38 AM UTC-5 Bruno Marchal wrote:

Philip Thrift

unread,
Aug 8, 2020, 6:56:57 AM8/8/20
to Everything List


This is the view of the physicist as a kind of religious mystic, who contemplates a physics outside of language, and some "truth" out there they will never find.

But all there is to write/speak with is language, 

e.g.

R_{\mu \nu} - {1 \over 2}g_{\mu \nu}\,R + g_{\mu \nu} \Lambda = {8 \pi G \over c^4} T_{\mu \nu}


 and how different vocabularies understood pragmatically might be translated into— or reduced to— one another

Bruno Marchal

unread,
Aug 10, 2020, 6:15:55 AM8/10/20
to everyth...@googlegroups.com
On 7 Aug 2020, at 16:19, Philip Thrift <cloud...@gmail.com> wrote:


Q. How does quantum entanglement occur in your models?

A. Two global Wolfram model states are said to be “entangled” if they share a common ancestor in the multiway evolution graph. Since spacelike-locality is not a necessary condition for branchlike-locality, it is possible for these states to be causally connected (i.e. to be connected in the multiway causal graph) even if they are not spatially local. This is the essence of quantum entanglement as it occurs, for instance, in the context of Bell’s theorem.



If that graph structure is the correct explanation of the entanglement, that graphe structure must be justified bt the observable-mode of self-reference, to be able to distinguish what is quanta and what is qualia.

Bruno




@philipthrift

On Friday, August 7, 2020 at 6:38:33 AM UTC-5 Lawrence Crowell wrote:
That might be, but a programming language that has no context with anything is not that valuable. At least it is not that valuable to me. My point is this seems to connect with concepts of spacetime as built up from large N entanglements.

LC

On Thursday, August 6, 2020 at 4:17:44 PM UTC-5 cloud...@gmail.com wrote:
Ultimately this is not really about "theory" (in the usual sense) at all, It is about defining a programming library: So it's really in the end a programmer's manual.

Wolfram Physics Project Functions


(an extension of the Wolfram language/ecosystem)

All of general relativity, quantum mechanics, and whatever comes next is to be written as programs in this library/language.

@philipthrift






--
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-li...@googlegroups.com.

Bruno Marchal

unread,
Aug 10, 2020, 6:29:37 AM8/10/20
to everyth...@googlegroups.com
On 7 Aug 2020, at 17:53, Philip Thrift <cloud...@gmail.com> wrote:


On Friday, August 7, 2020 at 9:43:38 AM UTC-5 Bruno Marchal wrote:
On 7 Aug 2020, at 13:38, Lawrence Crowell <goldenfield...@gmail.com> wrote:

That might be, but a programming language that has no context with anything is not that valuable.


At least, we should take the semantic of the reality on which that language is based. A language per se is not enough.





At least it is not that valuable to me. My point is this seems to connect with concepts of spacetime as built up from large N entanglements.

I agree, and the entanglement must be explained from the first person indeterminacy, singular and plural, which are imposed by incompleteness on all “creatures” living in arithmetic (or at its internal phenomenological border given by the self-reference mode available to the universal machine.

With mechanism, physics is a branch of machine biology (or psychology, or better “theology” …).

Bruno


But that has nothing to do with physics..

How could a derivation of physics have nothing to do with physics? You are right, physics describes some reality through number relation (like F = ma,  F = GmM/r^2, …), but with Mechanism, those physical (ov-bservable, locally predictable) number relations must be explained in a sort of Darwinian way from the “number”s dream” (which we know to be all realised in arithmetic).




Physics only describes (in a language).

Musing about its "meaning" is for philosophers to waste their time on.

With mechanism, physics becomes a branch of some philosophies or theologies, making them testable. 
Philosophy and theology are just nickname for fundamental science. People who do not muse on this usually accept the Aristotelian theologies, which unfortunately are logically incompatible with Darwin or any inferred physical laws until now, including the non computable part.




"Our best computer simulations, accurately describing everything [in physics], use only finite computer resources by treating everything as finite."
-- Max Tegmark


That is locally correct, and exploited through Mechanism (by Nature), but to get the qualia, there is not much choice to come back to Plato.

You can divide the theologies in three categories:

1) those with a Creator and a Creation,
2) those with only a Creation 
3) Those without a Creator and without a Creation

1) and 2) are what I called Aristotelian theologies. They “believed” in some “creation”, with or without a creator (personal or not).

3) is Plato (although Plato was just searching and looking at alternative, and we could add all the nuances between Platonism, Middle-Platonism (neopythagoreanism) and néoplatonisme.

Mechanism enforces the theologies of type three “3)", also called “neutral monism”. There is still a sort of God, but it makes not much sense to look at It like it could be a creator (the arithmetical reality, which supports all computations, from which the universal machine will infer an observable reality, a long time before realising it was also deducible.

Bruno







@philipthrift 

--
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-li...@googlegroups.com.

Bruno Marchal

unread,
Aug 10, 2020, 6:44:59 AM8/10/20
to everyth...@googlegroups.com
On 7 Aug 2020, at 20:35, Lawrence Crowell <goldenfield...@gmail.com> wrote:

Context is all if you are doing science, for in science we study objects and events.


I think science is more general than that. When you do metaphysics with the scientific method, it might be better to not postulate objects and events, as this seems to presuppose already Aristotelian theology.

Np need to military science. Science can study anything, propose theories about anything, as long as it gives mans of testing the theories, and evaluating their benefits.




If your interest is in doing pure mathematics or computer science that is fine, but it in of itself does not give physics.


Have you read my papers? I can prove that IF we assume Mechanism, then physics has to be justified entirely by the machine theology (by which I mean the study of the intensional variant of Solovay’s logic G*, as I have explained sometimes).




Feynman made some note of this. I found this little science fiction clip interesting along these lines. It is about a dormant computer system activating an attack sequence in a war that is long over. Note who in a sense "won the war." The machines activate algorithms with no context to reality.

Like brain and universal machine. Yes, they dream a lot, but from their own perspective, they belong to infinities of computations, and that is what we observe below pur substitution level. 
There is always some context with the basic reality, as a computation is a very particular number relation. You need a reality to have computations, but the physical reality is not an ontological reality: it becomes a first person plural observable by infinities of numbers. That is testable, and indeed it predicted both the “MWI” of physics, and the quantum formalism, at least up to now.
The evidences accumulated that the physical observable are the canonical observable of neopythagoreanism. 
In fact, there are no evidence for a primary matter or for physicalism. The Renaissance has been only half-enlightenment: science will resume when we will also doubt in the fundamental (philosophy, religion) domain. 

You can compare with EPR. When I was young I was told that I would waste my time in studying such philosophical papers, but Bell contradicted this already and Shimony understood that what is thought as belonging to philosophy can become science later, as both theory and experimentation are improved. Same here: mechanism in theology is completely testable (that is: refutable), so we will see, soon or later, if Nature contradicts Mechanism. The truth itself can never be known as such (provably in the Mechanist theories).

Bruno



Bruno Marchal

unread,
Aug 10, 2020, 6:49:58 AM8/10/20
to everyth...@googlegroups.com
On 8 Aug 2020, at 12:56, Philip Thrift <cloud...@gmail.com> wrote:



This is the view of the physicist as a kind of religious mystic, who contemplates a physics outside of language, and some "truth" out there they will never find.

But all there is to write/speak with is language, 

e.g.

R_{\mu \nu} - {1 \over 2}g_{\mu \nu}\,R + g_{\mu \nu} \Lambda = {8 \pi G \over c^4} T_{\mu \nu}


 and how different vocabularies understood pragmatically might be translated into— or reduced to— one another




You have always three levels:

1) a language (grammar)
2) a theory (axioms, formula, theorems)
3) an intended reality supposed to give sense to the theorems

By incompleteness, “3)” is never entirely definable (still less provable) in the language or in the theory itself.

With mechanism, we cannot avoid the need to postulate at least one universal machinery, and then, all the rest must be deduce from self-reference if we want get the correct distinctions and relationships between sharable quanta and the private qualia.

Bruno







On Friday, August 7, 2020 at 1:35:27 PM UTC-5 Lawrence Crowell wrote:
Context is all if you are doing science, for in science we study objects and events. If your interest is in doing pure mathematics or computer science that is fine, but it in of itself does not give physics. Feynman made some note of this. I found this little science fiction clip interesting along these lines. It is about a dormant computer system activating an attack sequence in a war that is long over. Note who in a sense "won the war." The machines activate algorithms with no context to reality.


LC



--
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-li...@googlegroups.com.

Lawrence Crowell

unread,
Aug 10, 2020, 4:25:52 PM8/10/20
to Everything List
On Monday, August 10, 2020 at 5:15:55 AM UTC-5 Bruno Marchal wrote:
On 7 Aug 2020, at 16:19, Philip Thrift <cloud...@gmail.com> wrote:


Q. How does quantum entanglement occur in your models?

A. Two global Wolfram model states are said to be “entangled” if they share a common ancestor in the multiway evolution graph. Since spacelike-locality is not a necessary condition for branchlike-locality, it is possible for these states to be causally connected (i.e. to be connected in the multiway causal graph) even if they are not spatially local. This is the essence of quantum entanglement as it occurs, for instance, in the context of Bell’s theorem.



If that graph structure is the correct explanation of the entanglement, that graphe structure must be justified bt the observable-mode of self-reference, to be able to distinguish what is quanta and what is qualia.

Bruno


I indicated in the above message on Aug 6, 2020, 1:04:48 PM (4 days ago) how these graphs might pertain to entanglements. This appears somewhat communsurate with what Wolfram is saying. 

I am not terribly worried about qualia, and at this time tend to shy away from invoking consciousness in QM.

LC

Lawrence Crowell

unread,
Aug 10, 2020, 4:54:00 PM8/10/20
to Everything List
On Monday, August 10, 2020 at 5:44:59 AM UTC-5 Bruno Marchal wrote:
On 7 Aug 2020, at 20:35, Lawrence Crowell <goldenfield...@gmail.com> wrote:

Context is all if you are doing science, for in science we study objects and events.


I think science is more general than that. When you do metaphysics with the scientific method, it might be better to not postulate objects and events, as this seems to presuppose already Aristotelian theology.

Np need to military science. Science can study anything, propose theories about anything, as long as it gives mans of testing the theories, and evaluating their benefits.




If your interest is in doing pure mathematics or computer science that is fine, but it in of itself does not give physics.


Have you read my papers? I can prove that IF we assume Mechanism, then physics has to be justified entirely by the machine theology (by which I mean the study of the intensional variant of Solovay’s logic G*, as I have explained sometimes).



I remember reading something of yours a couple of years ago. You might have to send me the paper with this development.

The equation between quantum states and units of information is through the von Neumann quantum entropy and its parallel with Shannon;s formula. Transitions between states by interactions are then in a way modeled as a sort of computation or algorithmic-like process. I am not particularly given to the idea the universe is an algorithm though.

LC

Brent Meeker

unread,
Aug 10, 2020, 6:43:54 PM8/10/20
to everyth...@googlegroups.com


On 8/10/2020 3:29 AM, Bruno Marchal wrote:
How could a derivation of physics have nothing to do with physics? You are right, physics describes some reality through number relation (like F = ma,  F = GmM/r^2, …), but with Mechanism, those physical (ov-bservable, locally predictable) number relations must be explained in a sort of Darwinian way from the “number”s dream” (which we know to be all realised in arithmetic).

A precis of an "everything" theory.  What we observe must be as predicted by the theory (otherwise it's invalidated). And the theory predicts everything (therefore it's useless, but can't be invalidated).

Brent

Philip Thrift

unread,
Aug 11, 2020, 5:35:52 AM8/11/20
to Everything List

"I am not particularly given to the idea the universe is an algorithm though."
- LC

What tbeory of theoretical physics today (GR, QM, ...) *cannot be replicated* as (simulation) programs running on supercomputers (like the ones at universities and national labs)?

What is a single example in physics for which this is the caae?

@philipthrift


Bruno Marchal

unread,
Aug 12, 2020, 3:38:49 AM8/12/20
to everyth...@googlegroups.com
On 10 Aug 2020, at 22:25, Lawrence Crowell <goldenfield...@gmail.com> wrote:

On Monday, August 10, 2020 at 5:15:55 AM UTC-5 Bruno Marchal wrote:
On 7 Aug 2020, at 16:19, Philip Thrift <cloud...@gmail.com> wrote:


Q. How does quantum entanglement occur in your models?

A. Two global Wolfram model states are said to be “entangled” if they share a common ancestor in the multiway evolution graph. Since spacelike-locality is not a necessary condition for branchlike-locality, it is possible for these states to be causally connected (i.e. to be connected in the multiway causal graph) even if they are not spatially local. This is the essence of quantum entanglement as it occurs, for instance, in the context of Bell’s theorem.



If that graph structure is the correct explanation of the entanglement, that graphe structure must be justified bt the observable-mode of self-reference, to be able to distinguish what is quanta and what is qualia.

Bruno


I indicated in the above message on Aug 6, 2020, 1:04:48 PM (4 days ago) how these graphs might pertain to entanglements. This appears somewhat communsurate with what Wolfram is saying. 

I am not terribly worried about qualia, and at this time tend to shy away from invoking consciousness in QM.


Where we assume Mechanism/computationalism (i.e.Descartes made mathematically precise through Turing), the quantum is explained from a theory of consciousness. It is the other way around than those who try to explain Consciousness from the quantum.

Actually, I found the MW by myself exclusively from mechanism (and amoeba’s observation), but at that time, I thought that I was refuting computationalism. It took me to discover Everett’s paper to realise that the physicists did already suspect that the physical reality might be multiple (which is just obvious with mechanism as long as you are OK that 2+2=4: that is already proved implicitly in Gödel 1931 paper).

Here, all you need is the characterisation of consciousness as something true, indubitable, immediately knowable, non provable and non definable (without invoking the notion of “truth”).

With this, all introspective machine discover “consciousness”, and the (unique precise physics) which go with it, and thank to the QM-MWI, we got striking confirmation of that physics. It took me 30 years more to get the quantum logic, and normally it is a “type 0” à-la von Neumann quantum logic, from which a unique measure and integral must be derived.

Physics becomes a subbranch of the theology/psychology of the universal Turing machine/number.

Now, this physics has no reason to replace the empirical physics, but its advantage is that it explain the quanta, the qualia, and their relation. It shows also that in theology or metaphysics, we need to come back to Plato type of theology. The Aristotelian are recovered as local approximation.

Bruno






LC

 
@philipthrift

On Friday, August 7, 2020 at 6:38:33 AM UTC-5 Lawrence Crowell wrote:
That might be, but a programming language that has no context with anything is not that valuable. At least it is not that valuable to me. My point is this seems to connect with concepts of spacetime as built up from large N entanglements.

LC

On Thursday, August 6, 2020 at 4:17:44 PM UTC-5 cloud...@gmail.com wrote:
Ultimately this is not really about "theory" (in the usual sense) at all, It is about defining a programming library: So it's really in the end a programmer's manual.

Wolfram Physics Project Functions


(an extension of the Wolfram language/ecosystem)

All of general relativity, quantum mechanics, and whatever comes next is to be written as programs in this library/language.

@philipthrift

--
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-li...@googlegroups.com.

Bruno Marchal

unread,
Aug 12, 2020, 3:59:30 AM8/12/20
to everyth...@googlegroups.com
On 10 Aug 2020, at 22:54, Lawrence Crowell <goldenfield...@gmail.com> wrote:



On Monday, August 10, 2020 at 5:44:59 AM UTC-5 Bruno Marchal wrote:
On 7 Aug 2020, at 20:35, Lawrence Crowell <goldenfield...@gmail.com> wrote:

Context is all if you are doing science, for in science we study objects and events.


I think science is more general than that. When you do metaphysics with the scientific method, it might be better to not postulate objects and events, as this seems to presuppose already Aristotelian theology.

Np need to military science. Science can study anything, propose theories about anything, as long as it gives mans of testing the theories, and evaluating their benefits.




If your interest is in doing pure mathematics or computer science that is fine, but it in of itself does not give physics.


Have you read my papers? I can prove that IF we assume Mechanism, then physics has to be justified entirely by the machine theology (by which I mean the study of the intensional variant of Solovay’s logic G*, as I have explained sometimes).



I remember reading something of yours a couple of years ago. You might have to send me the paper with this development.

I will do that.



The equation between quantum states and units of information is through the von Neumann quantum entropy and its parallel with Shannon;s formula. Transitions between states by interactions are then in a way modeled as a sort of computation or algorithmic-like process. I am not particularly given to the idea the universe is an algorithm though.


Mechanism makes this impossible. The physical universe is an emerging, in the mind of the universal machine, of a *non* computable first person (plural) pattern.

At each instant (indexical computational step) you have an infinity of computations (arithmetical concept) going through you state. To predict any first person experience (like seing a the position of a needle on some device), you need “in principle” to look at all computations going through that state, and that can be shown to be not computable a priori.

A slogan could be: IF I am a machine, then what I am not is NOT a machine. 

Most attributes of a machine (even non universal) are not computable/decidable. Already, the set of programs computing any functions is not a decidable set (Rice theorem), and no person-machine can know which machine support her, or which computations access her.
(This is easy to prove, but is also rather obvious if you thing to program factorial using some partially undecided subroutine).

I will send you paper soon. Meanwhile, you can consult my sane04 summary papers, 



and if your institution follows some journal, you might get them from herebelow:

Marchal B. The computationalist reformulation of the mind-body problem. Prog Biophys Mol Biol; 2013 Sep;113(1):127-40
https://www.ncbi.nlm.nih.gov/pubmed/23567157

Marchal B. The Universal Numbers. From Biology to Physics, Progress in Biophysics and Molecular Biology, 2015, Vol. 119, Issue 3, 368-381.
https://www.ncbi.nlm.nih.gov/pubmed/26140993

B. Marchal. The Origin of Physical Laws and Sensations. In 4th International System Administration and Network Engineering Conference, SANE 2004, Amsterdam, 2004.
http://iridia.ulb.ac.be/~marchal/publications/SANE2004MARCHALAbstract.html (sane04)

Plotinus PDF paper with the link:
Marchal B. A Purely Arithmetical, yet Empirically Falsifiable, Interpretation of Plotinus’ Theory of Matter. In Barry Cooper S. Löwe B., Kent T. F. and Sorbi A., editors, Computation and Logic in the Real World, Third Conference on Computability in Europe June 18-23, pages 263–273. Universita degli studi di Sienna, Dipartimento di Roberto Magari, 2007.
http://iridia.ulb.ac.be/~marchal/publications/CiE2007/SIENA.pdf
(http://iridia.ulb.ac.be/~marchal/publications/CiE2007/SIENA.pdf)

Marchal B. The East, the West and the Universal Machine, Progress in Biophysics and Molecular Biology, 2017, Vol. 131, pp. 251-260.
https://www.ncbi.nlm.nih.gov/pubmed/28919132

Marchal B.  Religion, science and theology, similarity and differences, Dialogo Journal, 2018, Vol. 5, pp. 205-218.
(available at http://www.dialogo-conf.com/archive/)






LC
 


Feynman made some note of this. I found this little science fiction clip interesting along these lines. It is about a dormant computer system activating an attack sequence in a war that is long over. Note who in a sense "won the war." The machines activate algorithms with no context to reality.

Like brain and universal machine. Yes, they dream a lot, but from their own perspective, they belong to infinities of computations, and that is what we observe below pur substitution level. 
There is always some context with the basic reality, as a computation is a very particular number relation. You need a reality to have computations, but the physical reality is not an ontological reality: it becomes a first person plural observable by infinities of numbers. That is testable, and indeed it predicted both the “MWI” of physics, and the quantum formalism, at least up to now.
The evidences accumulated that the physical observable are the canonical observable of neopythagoreanism. 
In fact, there are no evidence for a primary matter or for physicalism. The Renaissance has been only half-enlightenment: science will resume when we will also doubt in the fundamental (philosophy, religion) domain. 

You can compare with EPR. When I was young I was told that I would waste my time in studying such philosophical papers, but Bell contradicted this already and Shimony understood that what is thought as belonging to philosophy can become science later, as both theory and experimentation are improved. Same here: mechanism in theology is completely testable (that is: refutable), so we will see, soon or later, if Nature contradicts Mechanism. The truth itself can never be known as such (provably in the Mechanist theories).

Bruno

--
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-li...@googlegroups.com.

Bruno Marchal

unread,
Aug 12, 2020, 4:05:08 AM8/12/20
to everyth...@googlegroups.com
The theory (the physics derived from machine’s introspection) does not predict everything. It already rules out all classical physics, and predict a quantum physics with “many-histories”, and a non standard quantum statistics. It suggests a symmetrical important reversible core. So, on the contrary, it predicts something completely specific, but with a multiverse sort of shape. What we observed is exactly what is predicted, until now. 

Of course to do physics with the machine physics would be like to do a pizza using superstring theory. The goal is not a goal in physical science, but to get a theory of consciousness (and everything) compatible with the observations. 

Bruno




Brent

--
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-li...@googlegroups.com.

Lawrence Crowell

unread,
Aug 12, 2020, 5:10:47 PM8/12/20
to Everything List
On Wednesday, August 12, 2020 at 2:59:30 AM UTC-5 Bruno Marchal wrote:
On 10 Aug 2020, at 22:54, Lawrence Crowell <goldenfield...@gmail.com> wrote:




I remember reading something of yours a couple of years ago. You might have to send me the paper with this development.

I will do that.



The equation between quantum states and units of information is through the von Neumann quantum entropy and its parallel with Shannon;s formula. Transitions between states by interactions are then in a way modeled as a sort of computation or algorithmic-like process. I am not particularly given to the idea the universe is an algorithm though.


Thanks for these references. I have a bit of a queue of papers yet to read.

LC
Reply all
Reply to author
Forward
0 new messages