Dark energy as inertia of information-reference?

[Fra]

The purpose of this thread is not so much to discuss immature details but to see what the general opinon is about the suggestions that "dark energy" might be understood as a kind of information energy.

As far as I know, I have not myself seen any fully satisfactory papers on this, but there are some thinkers and papers that are sniffing in the direction.

M. Paul Gough associates Landauers principle of "heat dissipation" associated with releasing of a boolean state. He makes some elaborations in "Information Equation of State"
-- http://www.mdpi.org/entropy/papers/e10030150.pdf

I can very much associate to the idea of relating information with energy, and that loosing control of a microstate means releasing energy, but I don't quite like the details of that paper. There are probably several ways to do this.

This also faintly relates to questions I asked in the thread probing the logic of E-H action with a cosmological term - https://www.physicsforums.com/showthread.php?t=239414. The reflection I made is that the confidence in the "microstructure itself" contains energy, and loosely this may be related to the dark energy. I think also relates to the physical meaning of symmetry. Ie the "symmetry itself" implies information! This is why there is no material stuff we can idenfity with it, it's the reference itself. This is related to landauers principle, althouhg I consider his idea a kind of semiclassical version, that needs improvement.

Anyway, I hope you get the general idea. Regardless of version of the details, the question is what you people think of research going in this direction? Do you think it's good, or do you think it's misdirected? After all it's somewhat unconventional abstractions.

Does anyone know where the bleeding edge research in this direction takes place?

/Fredrik

 

[atyy]

Not sure if it's bleeding or bled:

The Implications of a Cosmological Information Bound for Complexity, Quantum Information and the Nature of Physical Law, P.C.W. Davies, http://arxiv.org/abs/quant-ph/0703041

 

[Fra]

Thanks for the link atyy. It looks like a conceptual paper rather than mathematical implementations, but it seems nice.

I will try to read this tonight. (I just skimmed the first paragraphs and his first section "what are the laws of physics", well expresses the problem we need to solve. The problem of how to a attach intrinsic physical basis to the "laws of physics", is exactly the problem. The "assymmetry" of information, where the laws of physics are fixed, and the rest is dynamical is very much unsatisfactory.(

/Fredrik

 

[friend]

The purpose of this thread is not so much to discuss immature details but to see what the general opinon is about the suggestions that "dark energy" might be understood as a kind of information energy.

Let's see, what do we know about information and dark energy. Information only exists in the context of probability distributions. And probability only exists when there is uncertainty. So... the uncertainty principle of QM gives rise to zero-point-energy which is equivalent to the cosmological constant (we hope) or a.k.a. dark energy. So dark energy should be equated to information.

 

[Fra]

Let's see, what do we know about information and dark energy. Information only exists in the context of probability distributions. And probability only exists when there is uncertainty. So... the uncertainty principle of QM gives rise to zero-point-energy which is equivalent to the cosmological constant (we hope) or a.k.a. dark energy. So dark energy should be equated to information.

Thanks for your comment. So we can agree so far!

Then the remaining question seems: What is the resolution to the absurdity that the missing information content in vacuum is ether infinite, or best orders of many magnitues off scale? This I think is more than just a detail.

Can that problem somehow be attacked also by some what "pure" information ideas? how?

/Fredrik

 

[friend]

Then the remaining question seems: What is the resolution to the absurdity that the missing information content in vacuum is ether infinite, or best orders of many magnitues off scale? This I think is more than just a detail.

Well, let's see, physical information has been associated with Horizions... recall the information lost in a black hole paradox, etc. So when talking about the universe, I guess the only relevant horizon would be the cosmological event horizon, the distance from us at which space is presently expanding faster than the speed of light. We are prevented from seeing that far at present because the farther we see the further back in time we see; so we never see what's happening now at those distances. But this horizon would mark the boundary within which lies the total information content of the universe at this time. Is this total information constant or changing? How would we answer that?

 

[Fra]

Well, let's see, physical information has been associated with Horizions... recall the information lost in a black hole paradox, etc. So when talking about the universe, I guess the only relevant horizon would be the cosmological event horizon, the distance from us at which space is presently expanding faster than the speed of light. We are prevented from seeing that far at present because the farther we see the further back in time we see; so we never see what's happening now at those distances. But this horizon would mark the boundary within which lies the total information content of the universe at this time. Is this total information constant or changing? How would we answer that?

Friend, I think you answered the original question as it seems that you like this direction! Thanks. There are many details left but that might be a topic for another thread I think.

I fell asleep last night so I quickly skimmed the paper atyy posted this morning.

In general I like the ideas of the paper, but it's clearly not very elaborated. One difference in my opinion relative what I can see in that paper is that - to continue his computer analogy - I would extend the analogy to not just say that the universal laws are a computer program running of physical hardware, I would say that any observer is a kind of "self-assembled" computer program, that is communicating with other similarly minded "self-assembled system". I think such a generalization is nesessary because in the big picture we do not only need to give an isolated answer to what dark energy might be, but also to explain the logic of the standard models of particle physics, and explain it's open questions and hopefully also some kind of unification. I have great hopes that this direction can not only answer some of the key questions, but that it can also provide possible road to unification.

This would in principle, by analogy mean that each observer has it's own "effecive" horizon (which of course doesn't IMO necesarily refer to ordinary spacetime), and it's own "opinion" or better called expectation of "physical law", and I think that this information processing idea can explain some of the other known interactions as a result of different opinion on "physical law". Two particles might be though to interact because they are "negotiating" on physical law.

Anyway, I think, if we really do believe in this direction, the foundations of QM needs to be reformulated. As the paper notes, the universe is so large that small deviations are not detectable, but in my analogy that each small observer are also like a small computer - inside the "big computer" - then the large number approxiation doesn't apply - things will happen, probably interactions will appear that are due to the discreteness of information. In particular would one expect such behaviour when it comes to ordinary particle physics. I think this connection - SCALING physical law, down to a very "small computer" - is what I wish to see someone work out.

/Fredrik

 

[Fra]

and it's own "opinion" or better called expectation of "physical law" , and I think that this information processing idea can explain some of the other known interactions as a result of different opinion on "physical law".

To make this explicit, in a mathematical model, is the key of what I am personally fighting with. They key is what does "expectation of physical law" really mean? I do not picture a frequency type definition of probability, I rather picture a sort of "logical probability", based on an evolved logic. Very much like the spirit of the inside-view of a game - a player set out to play for his life, and part of the game is to try to figure out the rules of the game, and by feedback, if he is successful he can even participate in making the rules. This very thing, I have found I think no papers on unfortunately.

/Fredrik

 

[Fra]

There is a way that the spirit of what I suggested here can be understood in a very intuitive way. Most of use has natural understanding of how sociology, economy, human pshychology, peace and war works, right?

There was once a businessman (forgot who) who once exclaimed that "Right or wrong? That is a matter of negotiation!"

That might sound like a cynical statement of someone whose objective is to make money without moral second thoughts. However I think it's in fact a basic observation about (like it or not) how nature seems to work.

Alot of human conflicts boils down to a very simple thing. Differing opinon on what is "right and wrong". And the result of this "inconsistency" is conflict, or "war" at worst.

There are many possible outcomes of the conflict. One may eat the other, or there may take place a sort of "negotiation", where both parties revises their opinon in the light of the opinion of the other one. A new equilibrium may from as a compromise.

We can say also say the "true or false" is a matter of reference. And this inconsistency gets consequences when mixed. The interaction is a result of the apparent inconsistency, and one way or the other, a new revised consistenct is typically formed.

A physical interaction? Maybe, maybe not?

This is a different way of thinking, as compared to the old ideals of realism.

Anothre trait of this abstraction, is that it is very general. Clearly it can be applied not only to physics, but also to economy, sociology conflicts etc.

What we need is a nice mathematical formalism that implements such systems. I'm not economist but I think similar looking system probably exists in economy (or it should, if I was an economist this is what I would do if it wasn't already done; which I'm sure it is). Now how does this really apply to physics? Where each part of the universe is "negotiating" with the other parts? And what does this insight, tell use about the foundations of QM, the problem of QG etc? My WAG is alot.

/Fredrik

 

[MTd2]

Fra, it's funny that I had this same reasoning for years, but no one to talk about it
The one guy with the papers you need is Raphael Bousso. It's from him that I learned and started thinking about these ideas. Try to surch for his articles on holography. It starts in 1999 and goes until 2004, more or less. After that, he got addicted in the multi universe ****.

Well, so, how on observer communicats with others? At the most minimum scale?

 

[Fra]

Hey MTD2, I'm glad that you connect to the ideas! I can't say I have anyone to talk to about this either. OTOH, I still have "internal equilibration" to do, and I consider that my personal problem. I have never tried to communicate my immature ideas in detail to anyone, it's too early for that. Meanwhile I am trying to read the work of others. I figure I have several years of personal work before it's worth convincing any opponents. Either you are on the same page already, or you are not, in which case you need to pull off a small miracle to get attenation.

I have read some papers of Raphael Bousso, and from what I remember he has interesting ideas. But I never recalled seeing anything that I perceived to be "dead on" so to speak, but I can give it another try and update myself on his papers.

> Well, so, how on observer communicats with others? At the most minimum scale?

I sure don't have the answer yet, but I think it's a pretty good question :) I am *slowly working* on translating these conceptual ideas into a mathematical formalism, and at the moment, the abstraction at hand is a discrete information model, which I think of as a "replacement" of koglomorov continuum probability. At this stage the concept of real number doesn't yet exists. From this I hope to construct "intrinsic measures" which I think of as microstructures or systems thereof. And these mictrostructures relate to each other, the relations corresponds to transformations. It's hard to explain this briefly.

But I am introducing a new meaning for A AND B when A and B refer to different spaces. This togetheter with some other "selection principles" could if I'm right imply quantum logic as an emergent (more fit version) of logic. In this the idea of inertia is also central.

Also this ides unifies the concept of entropy an the concept of action, via a concept of relative entropy. So the arrow of time should also be emergent as this elementary level.

/Fredrik

 

[Fra]

The one guy with the papers you need is Raphael Bousso. It's from him that I learned and started thinking about these ideas. Try to surch for his articles on holography. It starts in 1999 and goes until 2004, more or less.

I found 47 papers from Bousso or archive, and I saved them all in a folder. I'll try to skim them and see what I find.

One problem with connection I have with the cosmological focuses, continuum models etc is how to connect to the small - here I am currently thinking in discrete structures and emergent interactions constraints ONLY by complexity. I know of the covariant bound papers but I will try to skim also the other of his papers.

Thanks for your suggestion!

/Fredrik

 

[MTd2]

I am currently thinking in discrete structures

What do you think it is a discrete structure?

BTW, did you download them all manualy or automaticaly?

 

[Fra]

What do you think it is a discrete structure?

Difficult to motivate briefly I think. For me I've given this a lot of thought, as in given this ridicilous task I set out solve, where do I start?

Jaynes in his book, Probability theory - the logic of science, early in the book, makes the association that degrees of plausability is represented by a real number. I think that is jumped way ahead of things at that point. A real number is something fairly nontrivial IMO.

One of the first things I'd like to say though is that I don't think the difference between discrete models and continuum models are that large as it seems. Because I think that many of those to use continuum models, see a distinction between the mathematical continuum and a physical continuum. But I just doesn't see the advantage of thinking in terms of a redundant language, when most real life calculations on computers are truncating these anyway.

And the reason why real computers don'y use real numbers is also obvious, and leads to the other objection. To fully represent the spectrum of a real number, you need infinite memory. So it just don't make sense. And I think reality is economical. Simplicity. A real number is simple in thought, but when you think about making a real life computation on a real number it's not simple.

I try to shave off the redundancy of mathematical degrees of freedom and focus on what is in fact distinguishable. If we look at reality, and the way we do computations. In computers real numbers are always represented by rational ones.

This isn't just pragmatical computing economy, it also helps my brain. When you start from a concept of distinguishability, I have found that you naturally get first into basic logic, that is boolean logic, then comes natural and rational numbers, and the real numbers are a limiting scenario, which become effective when we are considering "large numbers".

But the idea of discreteness isn't IMO to just "approximate" the real numbers with a discrete lattic, that is totally the wrong idea I think. The concept of distinguishability, which I consider central to reconstructing an information concept, it naturally discrete.

Also the problem of choosing the prior in continuum models in bayesian probability, leaves us with a choice of infinite options - this makes no sense to me. I think that a fundamental continuum of choices lacks physical basis. This is why with the real number concept as a basis, you are lead to the normal probability theory. In which I see no natural measure of missing information.

So.. when trying to imagine how you would DISTINGUISH anything, any two states... I am lead to the conclusion that the continuum is a limiting case where the observer is infinitely complex, but that leads to the problem of stalled processing, unless the processing capacity is also infinite. In short the continuum reasonning as a fundamental starting point just leads to absurd situations.

So I am suggesting, like we do in mathematics, the real numbers are a kind of limiting case of rational numbers, not the othre way around. I think this order, should also be reflected in our physical models.

Others may disagree but no one has convinced me of how real numbers, are more fundamental than integers and rational ones. To me the opposite is intuitively close to obvious.

(I downloaded them manually.)

/Fredrik

 

[Coin]

The purpose of this thread is ... to see what the general opinon is about the suggestions that "dark energy" might be understood as a kind of information energy. ... I can very much associate to the idea of relating information with energy, and that loosing control of a microstate means releasing energy

So, what is confusing me about this suggestion is that I have always seen information associated with entropy. At the least, whatever mathematical definition of "information" you use it will require more information to describe something which is high entropy than something which is low entropy. This makes it seem like if dark energy is in fact just some kind of "energy cost" for information, then dark energy would be more intense in regions where there is high entropy... :/

If on the other hand we assume that dark energy is energy released by the erasure of information a la Landauer's Principle, then things seem even more confusing since the amount of entropy (and thus the amount of information) in the universe is in fact always increasing... right? (Do I have something backward here?)

Can you explain how the relationship between information and entropy is to be interpreted in the context of the kinds of proposals you link?

 

[MTd2]

If on the other hand we assume that dark energy is energy released by the erasure of information a la Landauer's Principle, then things seem even more confusing since the amount of entropy (and thus the amount of information) in the universe is in fact always increasing... right? (Do I have something backward here?

Not in this case, because entropy here is not the whole entropy of the universe, just what can be observed, that is, everything not behind of the cosmological horizon, which is an apparent horizon. So, we are not talking about an isolated system here, we're talking about an open system.

That's why Fra is talking about observers, because here, what is important it is what can be seen, and what can be seen in this case, is an open system.

 

[Fra]

So, what is confusing me about this suggestion is that I have always seen information associated with entropy. At the least, whatever mathematical definition of "information" you use it will require more information to describe something which is high entropy than something which is low entropy. This makes it seem like if dark energy is in fact just some kind of "energy cost" for information, then dark energy would be more intense in regions where there is high entropy... :/

If on the other hand we assume that dark energy is energy released by the erasure of information a la Landauer's Principle, then things seem even more confusing since the amount of entropy (and thus the amount of information) in the universe is in fact always increasing... right? (Do I have something backward here?)

Can you explain how the relationship between information and entropy is to be interpreted in the context of the kinds of proposals you link?

For myself, Landauers principle as it's stated is not something I am starting with. I see it more as an interesting reflection, that is onto something.

The way I picture this is way beyond erasing a computer memory. The computer memory isn't really my choice of analogy but I rather think of this closely connected to release of "random information", which is close enough to the least "useful" form of energy, such as head or blackbody information. I picture that suppsed a given observer has finite capacity to relate to it's environment - his memory is "full". What happens when he is continously beeing fed by new information? Then either he needs to grow larger memory, or he needs to release one storage slot to make room for the new information. Then a choice neesd to be made, what "physical storage bit" (let's just call it that without necessarily associating to computer circuits) can he throw away(loose), and yet loose a minimum of information content so to speak? Then one can imagine a sort of probability distribution of bits to be released, which is contains pretty much no information (IE the observer can loose this bit, and loose not skills). I think the "distribution" of the released information relative the emitter in this case,

Another motivation for me to look into this, is that the normally used definition of entropy is absolute because the microstructure is considered fixed. I think a more physical (~observable) measure of distinguishable disorder is necessarily relative. One also need to
reflect that hte choice of microstructure means adding information too. This is the similar idea that one says that we have two set of degrees of freedom, one is the usual matter degrees of freedom, the other is those beloning to spacetime itself. I am trying to generalized this, and build this into the very foundations of the new "proabilistic foundation".

It's similary to bayesian probability, but the usual form of bayesian probability is still in a sense aboslute, since the relations are absolute. I have come to my personal standpoint that there exists no static construction here, and this I want to exploiting to generate time evolution! Ie. the reason for time, is that there exists no static consistency. The solution to this IS evolution. But the evolution is constrained by "computations" with in a way can be thought of as the origin of inertia.

MTD2 is right that that abstraction I'm working on aims to be the proper inside-view. IE. the interesting part is distinguishable degrees of freedom. This is why in my radical view, it makes no sense to talk about fundamental degrees of freedom in the sense of observer independence. This alone implies indeterminism from the inside view. I do think that once this is understood, it will also shed light on the foundations of QM, which is really IMO, deeper, it will be a new foundations for relational information. And the key is the "intrinsically distinguishable" - here the real numbers are emergent only, they are not fundamental to me. This is why my thinking starts in a way that reconstructs the continuum along the way. I think the properties of the continuum, is in the details of it's construction. I think a lot of the absurd details of renormalization will go away if this works - or put differently the renormalization is "built in", and it may be that PART of the so called "renormalizations" are in fact physical processes, and requires TIME to be performed! This is totally lost in the mathematical treatment of it today, where these mathematical models seems to exists independent of hte physical world (this can't make sense can it?) Tihs is another point also well raised by Paul Davies in this paper http://arxiv.org/abs/quant-ph/0703041

His treatment is just reflections (not a full model yet), but I share a lot of what he wrote, especially in the introduction about "what is physical law".

/Fredrik

 

[Fra]

Can you explain how the relationship between information and entropy is to be interpreted in the context of the kinds of proposals you link?

Almost needless to say but to point out that the ultimate motivation of all this - to others, will be when you can see a model implementing constructed in this spirit, take off and make a difference to some of the open questions in physics. I am not there yet. Until then I can only try to convey my way of reasoning, for others to like or not like.

Mm 'll try to elaborate more but don't expect any clear answer, I have none...

Since I picture the notion of "observed distinguishability" as a starting point, these are the fundamental degrees of freedom that the observer can RELATE to and MEASURE. I consider this startpoint in the senese of zurek's sentiment that "what the observer knows and what the observer is, is inseparable". The deep meaning of this IMHO at least (I can't speak for Zurek) is that an observer can not "measure" in terms of more fundamental DEGREES, something that defines his very existence. This doesn't mean that there aren't options in the environment, it just means that these options are not distinguishble from the inside view.

So, information (ie not a MEASURE of information) refers to just distinguishable states.
Which ultimately are represented by numbers. In the simplest case, integers, that simlpy INDEX the distinguishable states. These numbers are only "labels" and any other labelling systems like a, b, c would do as well. But I envision that any labelling system at this point is ismorphic with plain numbers.

Next comes the idea of "counting evidence", which for each index, has a counter. Now this yields an array of integers. But the entire structured is bounded, by a complexity which I consider to be one of the qualifiers for an observer. This can loosely relate to say "mass" or "energy", although the connection to spacetime is yet not explicity.

So we have not a discrete probability distribution, which is really more like permutations. Now the question appears, when this structure is fed with data, a new sample needs to replace an old sample. Also other options that may exists is if there is a more optimum way to represent this data? for example empty index slots makes not sense, so events that doesn't happen they disappear from the "index field" (associate to borel's law). But what if there are spontaneous transformations of the "index field" (althouhg it's not an "field" in the ordinary sense) and the counts, for example Fourier transforms etc. We can transform these "mictrostructurs" to other systems of mictrostructures, and some of these will - for certain data streams - be found to be more efficient.

Anyway, there must also be a way for an observer to grow more degrees of freedom. This can happen IMO if the observer develops optimal internal structure which allows him to "predict" or "correlate" with the datastream itself - ie to become one with the datastream, and now the point is almost the "opposite to borel's law, that when a guess becomes sufficiently "probable" (don't take this literally to suggest I'm doing ordinary probability theory) then it comes true, or the main point, indistinguishable from the truth, as by the inside observer. This is the mechanism that may (way further up in the evolution of these abstractions) be a key to understanding GRAVITY as a kind of "attraction" betwen observers!

The idea of entropy is to be a "measure of missing information", as in the degree of freedom of the microstates of a microstructure. IMO, it follows from this that in the larger picture (universe with lots of observers), it is not possible to find an universal measure of missing information. There are in fact a neverneding list of possible measures of information. As to which one is "right", is a matter of communication of the collective. I think what is de facto a somewhat universal entropy in physics, is a result of evolution of the universe. After all, the universe today is very ordered, so the apperance of universal laws, and references is a matter of billions of years of negotiation. It would be almost offensive if some kind of common reference had not emerged - the real intersting questions is, did the universe have much of a choice? One would suspect that at minimum the choices was constrained.

There many such entropy measures, some are just flat out defined, some entropies are "constructed" and argue to be unique measures following certain desired properties, for example the cox axioms etc. However the measure itself, is dependent on this *choice*. So I have never seens a convincing argument for the ultimate measure. And I think this isn't a coincidence at all - I think it doesn't exists in any meaningful sense except in a FAPP sense.

What I suggest is that there exists no universal unique measure of missing information. There are only relative measures. So in my view, there are only relative entropy, but not even this relation is absolute. This results (as I see it) in an instability, this is the key to understanding time IMHO.

OTOH, to answer the question, I think the ultimate entropy, while relative, for any given observer, this measure is given by the observers makeup. There is (in a certain sense) a unique optimum measure for each perspective. This determines (In my speculative opinon that is; to be shown) the reaction of the observer to the "datastream" from the unknown. But the observer doesn't KNOW in advance which is righ, it's merely doing a sort of random walk in the space of measures. This datastream DRIVES the evolution of the observer. This is why I don't think one can construct the ultimate model without experimental contact. The model is living and motivted only in the context of interaction with experiment.

The "experiment" of these abstract observer concept IMO, is this "data stream", whose bandwith is determined by the size of the index field. I'm working on relating this to the holographic bound. But the problem is that there are for sure limits, that's not the problem. the problem is to identify the actual 4D spacetime. And that will require fixing the othre details first mentioned above. I don't think there are any shortcuts in the sense of estimating the number of bits in the universe by thumb to do this. It's probably all in the details unfortunately.

This is why I personally think it's necessary to start building these traits into a new foundation of information.

/Fredrik

 

[Fra]

So, what is confusing me about this suggestion is that I have always seen information associated with entropy.

The simlpe answer to this is also that what I was trying to probe and convey in this thread is non-standard. Most of those others who has written papers on this admit that it's controversial.

Another comment on this: What I think is that the matter is more complex than a universal measure of information. For example the shannon entropy contains an implicit ergodic hypothesis. IE. the definition relates to a underlying mictrostructure (may it be physical memory device or whatever), what I was suggesting is that two observers, may have differing opinons on this mictrostructure. Thus one might need a kind of "compare parallell transport on riemann manifolds" transportation of the measure itself, to the othre observers. But as I elaborated on in the later responses below (which I suspect no one may undstand except MTD2 who seems to at least partially conect) is that this "transport" is not a simple one-2-one transfomration. The transformation iself is like a "computation" and involves evolution (time).

I think some problems, including the problem of time, originates from trying find simple static relations. Thsis is also what I meant with connecting entropy and action. They are both measures, that in this pictures more or less conincide, and the principle of maximum entropy and the principle of least action is pretty much the same thing. But one need the right abstraction to see the similarity.

/Fredrik

 

[MTd2]

GRAVITY as a kind of "attraction" betwen observers!

What gives mass to the observer?

 

[Fra]

What gives mass to the observer?

This is one of the keys of course and I have no clear communicatable answer yet, what I have is a reasoning. And I think almost by construction anyone who is looking to patch the current models will find this highly speculative. But then it's my speculation, I don't ask anyone else to take risks for me. I accept my own risks to develop my own ideas further.

While some of the ideas are contained in the reasoning I tried to described, the full answer is really in a certain sense analogous to Wilczek's idea of building mass without mass, as beeing a result of a constraint on massless parts. Mass can be "generated" as a result of evolution, starting from what I currently think of as an minimal disturbance. Disturbances can evolve, and some disturbances happen to be preserved, and grow (usually at the expense of fellow observers in the environment).

The real solution of this generation of mass, is analogous to generation of certainty out of uncertainty. IE. the mass generation is not too alien from the concept of learning by a kind of induction. An intuitive idea of WHY mass is generated is like... mmm suppose there is an itelligent learning algortithm, that is so cleverly constructed, that it can learn more from it's worst case mistake than it can loose. Ie. you step one inch back, but move the floor more than one inch forward - or at least to a probability that is high enough to be observably indistinguishable from the truth. This is an inductive relation between microstructure evolution, and microstate evolution (relative tot the mictrostructure). This induction, which closely resemebles circular reasoning at first glance, should yield a net progression.

I don't know yet how the best final model will look like. To be honest right now it looks more like a kind of self-adjusting algorithm (basically a self-modifying program; literally a living piece of software), because it will not be possible to find a fundamental form of say a diff equations. however, from application of that algorithm, i am sure that FAPP-type of improvements to the standard form of the laws of physics, say it's differential equations etc will be possible.

/Fredrik

 

[Fra]

An intuitive idea of WHY mass is generated is like... mmm suppose there is an itelligent learning algortithm, that is so cleverly constructed, that it can learn more from it's worst case mistake than it can loose. Ie. you step one inch back, but move the floor more than one inch forward - or at least to a probability that is high enough to be observably indistinguishable from the truth. This is an inductive relation between microstructure evolution, and microstate evolution (relative tot the mictrostructure). This induction, which closely resemebles circular reasoning at first glance, should yield a net progression.

I just read my own description and I just saw that the most important thing wasn't mentioned. The trick I envision is this: I do not a such "assume a net progression" (net progression btw, is analogous to constructivity) I just not that either there is progression, or there is not, there can be status quote and the mass is stable, or there can be loss of mass. And now the point is that the mass is the qualifier of the observer, and an observer that consistently looses mass, eventually dissappears. So there we see a sort of darwinism. Only observers that evolve and preserved this so called "constructive trait" are selected. This means that observers that disrespect this - although technically not banned, are simply highly unlikely to populate the world.

Not sure if that helped. But that additional thing is an important key in this reasoning. The whole idea involves unconstrained variation, but the selection principle only preserves/selects to a higher extent certain kind of constructive behaviour. It would be a self-contradiction to picture evolution and emergence of inconsistency. So it's a kind of self-enforcing logic.

/Fredrik