Relative information at the foundation of physics by Carlo Rovelli

Hi Carlo,

Thank you for your reply.

Yes, the arithmetic is correct.

No, it was fairly easy to find these formulas, it is all simple logic.

Have you checked out the proton's diameter and the proton's mass formulas?

If I am correct, wouldn't this give us a big clue about the current proton's radius measurement problem ?

A

Patrick

Dear Carlo,

Re your answer (28th July) to my post: doesn't your determinism mean that all future physical outcomes are already fixed and "set in stone"? (Obviously I'm not referring to the issue of complexity, which means that future physical outcomes in real life usually cant be predicted).

Cheers,

Lorraine

Dear Carlo,

Re your answer (28th July) to my post: doesn't your determinism mean that all future physical outcomes are already fixed and "set in stone"? (Obviously I'm not referring to the issue of complexity, which means that future physical outcomes in real life usually cant be predicted).

Cheers,

Lorraine Ford

Dear Prof Rovelli,

very interesting essay. I completely agree with you about the role of information in physics. It is the relation between the objects which counts.

Your two principles (leading to quantum mechanics) are not totally new. von Weizsäcker also discussed the abstract derivation of quantum mechanics from finite Hilbert spaces (like C^2). I discuss it partly in my essay.

Maybe one point of disagreement: In my opinion, the relation between the discrete structure of spacetime (like the area quantization of LQG) and the smooth spacetime is not well enough understand. Is it really necessary, that the spacetime itself is discrete? Using diffeomorphism invariance, the information contained in a smooth 4-manifold is discrete (as I explained in my essay). Also the substructures like surfaces are partly regid (see for instance a incompressible surface in a 3-manifold which can be a topological expression for area quantization). Do not misunderstand me, I'm not an oponent of the discrete spacetime but the relation between discrete and smooth is not really explored (in dimension 4 or lower any triangulation (or PL structure) is uniquely determed by a smooth structure and back).

Maybe you have time to look in my essay.

Best wishes

Torsten

Carlos,

Interesting. Can you explain what a 'virtual photon' is, or why Snell's Law fails at Maxwell's near'Far field transformation along with Fresnel's refraction? We have to call the unexplained non-linear effect 'Fraunhofer' radiation and slip it under the carpet to be rationalised later! Unless you can offer an explanation?

Do you not believe that understanding the underlying quantum mechanism to the transform will aid our comprehension, so remove the anomalies and paradoxes (including such as Kinetic Reverse Refraction, another well known effect from optical science with not consistent theoretical explanation).

A mathematical identity is of course not a true or even absolutely precise 'explanation' of natures uncertainties, just a good approximation, Yes?

So I am suggesting perhaps these are the things we DON'T yet fully understand, or do you truly believe we do?

Peter

Dear Carlo,

Thank you for a most illuminating essay, which deserves top marks.

You explained how "selected structures define finality". There is clearly a Lagrangian cast to this statement if the action is bound globally and is time symmetric. It appears that the relative information gained by one system (i.e., an observer) about another system (i.e., a quantum state) is reflexive. In other words, the act of measurement defines both systems. The observer erases the entanglement information which encodes the global interconnectedness of herself and the other system. Both she, as perceiver, and the measured state, as the perception, emerge as ontic entities. (See my essay "A Complex Conjugate Bit and It".)

This "infinite game of mirrors reflecting one another" brings to mind the image of Indra's net, with a jewel glittering at every node.

Best wishes,

Richard

Dear Professor Rovelli,

While you tell an enjoyable tale and make a good point; I found your essay a bit disappointing, playing as it does upon the common misconceptions about entropy, and the confusion that exists about the extent to which the different types of entropy are interchangeable. Of course; the Scientific American editors did it too, when they ran an article about non-linear entropy in the mesoscale regime by J. Miguel Rubi, by proclaiming on their cover that his studies show how nature breaks the 2nd law of thermodynamics. However; if you read Rubi's version, he carefully explains that thermodynamic entropy steadily increases while the system displays alternating regimes of chaotic and orderly forms - in apparent defiance of the 2nd law. Or at least that defiance is what appears true for those who equate entropy with disorder.

Personally; I prefer the spreading metaphor championed by Leff, for thermodynamic and mixing entropy. It seems that due to efforts by Lambert and others, the 'entropy is disorder' description is almost completely absent from Chemistry textbooks, as it brought confusion and was distracting from the thermodynamic aspects of Chemistry, but in Physics this idea of energy dispersal being associated with thermodynamic entropy was kind of skipped over as statistical models based on information entropy were broadly adopted or already in use. Still; it is arguable that the specific meaning defined by Clausius was already changed somewhat by Boltzmann, when he suggested that disorder's increase was associated with growing entropy, and that not all entropy is the same.

Also; I think it is often glossed over that there is a specific n involved, the number of molecules in molar volumes of gases, which has the effect to make the system under study quantum mechanical. When there is a discussion of microstates, it is in terms of equivalent microscopic state changes which are all equally possible because they yield the same macroscopic condition. It is as though nature is performing one mother of a path integral, while we are just reading the temperature and pressure on a gauge. I think nature's way of defining things is indeed through relative information, and you have done a wonderful job of conveying this. A constrained system could consist of cranks and pistons, but it could be a pair of entangled particles where one has been partially measured as well. So your essay was enjoyed, and it made me think; but confusion about entropy is a bugbear for me, and I think your conclusion stands just as well without confusing the issues.

All the Best,

Jonathan

Please note; this comment was first posted here last night, but it disappeared with technical problems today. I have slightly edited the copy, which I respectfully submit for your attention Carlo. I respect your work greatly, and hope not to offend, but I want it known where I stand on the Physics involved - or the description thereof.

All the Best,

Jonathan

Dear Carlo Rovelli,

It was good to highlight that what was important are relative entropy and relative information, needing a reference template or a reference physical system. For experimental physicists this is obvious for ALL measurable quantities - that is why they create 'standards of measurements', a stable and universal metric of measures, for basic as well as secondary quantities. All measurements and hence new quantitative knowledge is a relative measurement. Information is no exception since it is encoded in physical states.

I felt that the initial example of Colored and Charged balls and inadequate observers perhaps was not appropriate since one assumes good measurement or observational apparatus BEFORE an observation is claimed - otherwise a specific observer's handicaps are to be folded in (convoluted/traced) into the measurement, as some blur function or coarse graining, or tracing over some variable. So, while it is true that special initial states are required for apparent irreversible behaviour, the statement about irreversibility assumes that the observer to which the sytsem couples is genuine and not blind, not zombie and so on. Irreversibility is relative with respect to initial states, but the observer's handicaps should not enter its formulation in physics.

Unnikrishnan

Carlo, as you may be aware, FQXi has just been moved to a new server and many people have found that posts are missing. A post I sent to you on Aug. 1, 2013 @ 14:12 GMT (in reply to your post on Jul. 30, 2013 @ 12:34 GMT) is missing, so I am resending the post in case it never gets restored:

Carlo,

thanks very much for your reply, but I must disagree with your reasoning.

I would contend that there is no dividing line between "the underlying physical" reality and the "high level" reality. There is a difference in complexity, but there is no difference in the essential nature of reality between the high level and the underlying level. No new properties can miraculously appear ex nihilo at the "high level" that do not have foundations in "the underlying physical" level.

So if the underlying reality is completely deterministic, then the "high level" is exactly the same. Determinism means that (due to laws of nature) only one physical outcome is possible for each next moment in time. That is, at the high level, no "choice and responsibility" is possible for human beings in a deterministic universe. Surely, in a deterministic universe, to believe that "choice and responsibility" is possible is to deceive oneself.

Cheers,

Lorraine

The above post was from me. I was logged out in no time at all!!!

Lorraine

Professor Rovelli,

As the eminence grise of these conversations, may I ask why physics favors the concept of information, over that of energy?

As living organisms, we are the result of billions of years of evolution. The

consequence of this process is two fairly distinct systems. One is the central nervous

system, to absorb, organize and act on information. The other, the respiratory, digestive

and circulatory systems, serve to consume and process energy. So we exist as

manifestations of this dichotomy of energy and information, as medium and message.

Since energy is conserved, old information is erased in the creation of new information. This gives rise to the "arrow of time."

We think of time as sequential events and the basis of cause and effect. Otherwise known as narrative and logic. The assumption these are effectively one and the same is flawed though. Does yesterday cause today, anymore than one rung on a ladder causes the next?

Cause and effect is due to the exchange of energy, usually significantly altering any information it may have conveyed. In the instance of days, the sun shining on a rotating planet creates the sequence called days, where it is not some point of the present proceeding through the sequence, but the changing configuration of the energy causing the sequence to form and dissolve, thus future becoming past. Tomorrow becomes yesterday, rather than anything other than a personal or subjective vector from yesterday to tomorrow.

While physical laws determine the outcome of any particular situation, the physical input into any event is only completed by the occurrence of the event, therefore precluding full knowledge of any event to a particular frame. Not only that, but as the past is no longer physically real, any events were often a matter of subjective perspective and any such perspective changes with time, the past can only be said to have occurred, without a truly definitive, objective knowledge of what did occur. So the notion of determinism is epistemic at best.

And the sanctity of information does seem based on this assumption of some form of "blocktime" determinism.

One would think that if time were a vector from past to future, the faster clock would move into the future more rapidly, but the opposite is true. Since it ages/burns quicker, it recedes into the past more rapidly.

Time and temperature are like frequency and amplitude; Descriptions of the energy.

I think we need to get back to basics.

Regards,

John Merryman

Carlo,

thanks very much for your reply, but I must disagree with your reasoning.

I would contend that there is no dividing line between "the underlying physical" reality and the "high level" reality. There is a difference in complexity, but there is no difference in the essential nature of reality between the high level and the underlying level. No new properties can miraculously appear ex nihilo at the "high level" that do not have foundations in "the underlying physical" level.

So if the underlying reality is completely deterministic, then the "high level" is exactly the same. Determinism means that (due to laws of nature) only one physical outcome is possible for each next moment in time. That is, at the high level, no "choice and responsibility" is possible for human beings in a deterministic universe. Surely, in a deterministic universe, to believe that "choice and responsibility" is possible is to deceive oneself.

Cheers,

Lorraine

Dear Professor Rovelli,

While you tell an enjoyable tale and make a good point; I found your essay a bit disappointing, playing as it does upon the common misconceptions about entropy, and the confusion that exists about the extent to which the different types of entropy are interchangeable. Of course; the Scientific American editors did it too, when they ran an article about non-linear entropy in the mesoscale regime by J. Miguel Rubi, by proclaiming on their cover that his studies show how nature breaks the 2nd law of thermodynamics. However; if you read Rubi's version, he carefully explains that thermodynamic entropy steadily increases while the system displays alternating regimes of chaotic and orderly forms - in apparent defiance of the 2nd law. Or at least that defiance is what appears true for those who equate entropy with disorder.

Personally; I prefer the spreading metaphor championed by Leff, for thermodynamic and mixing entropy. It seems that due to efforts by Lambert and others, the 'entropy is disorder' description is almost completely absent from Chemistry textbooks, as it brought confusion and was distracting from the thermodynamic aspects of Chemistry, but in Physics this idea of energy dispersal being associated with thermodynamic entropy was kind of skipped over as statistical models based on information were broadly adopted or already in use. Still; it is arguable that the specific meaning defined by Clausius was already changed somewhat by Boltzmann, when he suggested that disorder's increase was associated with growing entropy, but that not all entropy is the same.

Also; I think it is often glossed over that there is a specific n involved, the number of molecules in molar volumes of gases, which has the effect to make the system under study quantum mechanical. When there is a discussion of microstates, it is in terms of equivalent microscopic state changes which are all equally possible because they yield the same macroscopic condition. It is as though nature is performing one mother of a path integral, while we are just reading the temperature and pressure on a gauge. I think nature's way of defining things is indeed through relative information, and you have done a wonderful job of conveying this. The constrained system could consist of cranks and pistons, or it could be a pair of entangled particles where one has been partially measured.

So your essay was enjoyed, but confusion about entropy is a bugbear for me, and I think your conclusion stands just as well without need for confusing the issues.

All the Best,

Jonathan

Dear Professor Rovelli - I am always inspired by your papers (and books); this essay is no different. Your review of entropy is very enlightening. It was perhaps an accident that I read your paper back to back with the essay by Christian Corda, in which he says something very similar:

[from Corda] "In principle, a process in which pure states evolve in completely mixed states does not contradict the laws of quantum mechanics because the apparent information loss is instead hidden by quantum entanglement. The term entanglement means that the quantum state of a quantum system composed by two (or more) subsystems depends on the quantum state of each subsystem even if they are spatially separated. When one sums up the information in the two subsystems the result will be less than the information in the original system. The apparent information loss results hidden inside correlations between the subsystems."

My favorite paragraph form your paper was: "Thus, the systems formed by atoms have necessarily information about one another in the sense of Shannon. The negative information (in this sense) that a system have about another is precisely the relative entropy of the second, which is relevant for the interactions with the first, and is the conventional thermodynamical entropy."

In my essay I introduced an idea that might first seem absurd. A trapped photon bouncing back and forth between two atoms will trap energy/information, and for all intents and purposes will be "dark", i.e. outside of (classical) time. I introduce the concept of subtime to describe relative reversible evolution between measurements; much of this was inspired by your earlier writings.

I think you will find this idea to be quite different to thermal time and Tomita flow. I would be honored to receive your thoughts after reviewing it.

Kind regards, Paul

Hi Carlo,

Thank you for such a clearly described investigation into the fundamentals. I especially like these:

> If they are like letters of an alphabet, to whom do they tell stories?

> The world is not just a blind wind of atoms, or generally covariant quantum fields. It is also the infinite game of mirrors reflecting one another formed by the correlations among the structures formed by the elementary objects.

In my essay Software Cosmos I show how we might work within the simulation paradigm to model a world of information. This discrete computational cosmos has properties that we observe of our own, and in fact, I have carried out a test to see whether we now inhabit such a world.

Interestingly enough, a world constructed of information has room for more than the "blind wind of atoms". Different architectural layers of the software cosmos can have quite different properties. The layers of Life and Mind can lie below the physical layer of Matter, and not be (as conventionally assumed) emergent from it. Life and Mind, then, can animate Matter, and so produce the phenomenology of consciousness.

I hope you get a chance to read my essay, as I would love to know what you think of my model.

Hugh