Another classical system that violates CHSH-Bell

[Ben6993]

Despite supporting Joy on this issue, I do not see anything that is bogus about Gill's work. And there has been no conclusive proof for Richard that he is wrong. He still believes, quite sincerely I am sure, that he is correct. I do not understand why the word "bogus" is often used here. Well maybe for psychological reasons which must surely be against the CoP, but not for physics reasons?

Fred might think that Richard's insistence on using flatland algebra on observed/laboratory results is bogus, but I cannot see that. I assume that, because QM sets laboratory results in normal 3D space, then Richard insists on doing normal algebra on lab results. It still needs to be proved that normal laboratory space is not adequately covered by 3D space, ie it needs results for Joy's exploding balls experiment. If other macroscopic experiments are already doing the equivalent job, then that is excellent and may soon resolve that issue. Calculating on hidden variables in multidimensions is OK as the hidden variables are not observables and are therefore not subject to 3D algebra.

The missing data need to be explained. I have a very poor mental image of the GA space. Like, say, spacetime without any matter in it and very static. If the GA space was more dynamic that might affect the sampling process needed in a simulation to describe space? Jay Yablon at the old science.physics.foundations has a new draft paper which includes a section on Bell. (I wish I understood Jay's paper better but I would need n [>2] decades of study.) If vacuum fields played a major role in the quantum correlations, then that would presumably be a dynamic space. I am just suggesting that uniform sampling suggests to me a static space whereas non-uniform sampling might be needed for sampling a dynamic space.

The higgs fields can be weak isopin of either -0.5 or +0.5. I see these as potentially background spaces with - twist or + twist similar to GA's trivectors. But that is very speculative and likely to be incorrect.

Other questions are why are the hidden variables not afected by zbg or interactions with the higgs during time of flight. And what causes the quantum correlation for photons which do not have the 4pi periodicity?

[Joy Christian]

Despite supporting Joy on this issue, I do not see anything that is bogus about Gill's work. And there has been no conclusive proof for Richard that he is wrong. He still believes, quite sincerely I am sure, that he is correct. I do not understand why the word "bogus" is often used here. Well maybe for psychological reasons which must surely be against the CoP, but not for physics reasons?

Ben,

Do you think that Gill's "bogus" challenge accurately reflects my experimental proposal, or do I have a point when I object here to his misrepresentations?

In any case, what is wrong with the 2N vectors explicitly produced in this simulation? Did he not claim that it was impossible to generate such 2N vectors?

[FrediFizzx]

Fred might think that Richard's insistence on using flatland algebra on observed/laboratory results is bogus, but I cannot see that.

Ben, that is not what I said at all!!! Gill wants to use flatland algebra in the simulation. That is what is bogus. In the real experiment, the correlations are calculated exactly like Joy says in his paper. In the simulation, you have to use something that models what the 3-sphere geometry is doing. IOW, you have to simulate what we think Nature is doing. It is quite simple actually.

[FrediFizzx]

On this one I agree with Jaynes and Bell.

I'm sticking with Einstein; non-local correlations are physical nonsense.

And Gill's award is not bogus, nor is that the issue here. Since Bell's theorem became widely known, people have been arguing about it without anything convincing coming out of it. It would be a breakthrough if someone could provide a simulation program that disproves Bell's theorem according to an independent jury such as the one Gill and Christian already provided; and that should be not too difficult (although still hard work) if these papers say what we think that they are saying. Inversely, if no such simulation program can be produced based on these papers, then we have to conclude that we misunderstood what they are saying.

You brought it up so Gill's challenge is an issue here now. See my reply to Ben. Richard doesn't seem to know the difference between a real experiment and a simulation of such experiment. Richard knows that his challenge is "rigged". Joy (and minkwe and others) have already produced successful simulations. It is your problem if you don't want to accept that. IOW, I doubt that anything would change your mind at this point if you can't see that the simulations are already successful.

Anyways, the main issue for this thread is that we have two recent classical experiments that show strong correlations. This gives credence that Joy's mechanical experiment should be done. And the simulations support that also. The only way we will find out if Joy is right or not is to do the experiment or something similar to it.

[Gordon Watson]

On this one I agree with Jaynes and Bell.

I'd welcome clarification here: What is being agreed to? Thanks.

[FrediFizzx]

On this one I agree with Jaynes and Bell.

I'd welcome clarification here: What is being agreed to? Thanks.

I suppose he is talking about "socks". But that has never been an issue as far as physical correlations go.

[Gordon Watson]

Jay Yablon at the old science.physics.foundations has a new draft paper which includes a section on Bell.

I'd welcome the link to this paper! Thanks.

[FrediFizzx]

http://jayryablon.files.wordpress.com/2 ... mplete.pdf
Though I didn't see anything specifically about Bell. ???

[Yablon]

It is more a quantification of Bohm, than about Bell, unless you take the view that Bohm is also about Bell.

Chapter 20 of http://jayryablon.files.wordpress.com/2 ... mplete.pdf.

Jay

[Ben6993]

Joy wrote:

Do you think that Gill's "bogus" challenge accurately reflects my experimental proposal, or do I have a point when I object here to his misrepresentations?

Hi Joy, There have been a number of challenges and I am not clear as to the exact one here. You link it to your experimental proposal so it must be a simulation of your macroscopic experiment. Your proposal specifies one correlation to be produced (equation 94). I do not see where it says in your proposal that there are to be no replications but there would seem to be no reason other than savings in cost, time, resources etc to use replications in a real experiment, and of course there are no cost savings in a simulation.

There can only be one reason to introduce replications in a simulation and that is to ensure a failure to pass a CHSH limit. However, Richard has been open [= genuine and not bogus] about two types of challenge. In one challenge he said there is no hope of beating the CSHS limit and in the second type there is a hope [and because of his new paper he seemed to be more interested in the second type]. And I agree with Michel that using a CHSH limit with too few degrees of freedom [eg using replications] ensures failure. However, there is no replication in the second type of challenge, even in a CSHS test using only four pairs of angles.

Fred says that Richard wants to use flatland correlations in the simulation. (You may have to correct me here ... ) Equation 94 uses flatland algebra? Where does the GA enter the calculations in the macro simulation? I am confused here. Part of my misunderstanding is that I can only see the case for micro effects at the moment and am yet to be convinced about the macro effects. I don't think that Richard has anything against using GA in the simulations so long as observables[= result at the detectors] can be manipulated using normal algebra. The next point is something only I seem to worry about [so I must be wrong] Because of the 4pi periodicity of the detector magnets, the detector angle should really be a hidden variable. Especially when experimenters are said not to distinguish between 360 + 5 degrees and 5 degrees (because of convenience in rotating heavy magnets). But in the experiment there are cameras, not magnets, and there cannot be a 4pi variability with camera images? So there is not much scope for GA in the simulation calculations for macro bodies once the file(s) of data has(have) been produced???

Joy wrote:

In any case, what is wrong with the 2N vectors explicitly produced in this simulation? Did he not claim that it was impossible to generate such 2N vectors?

I am not fluent with R language yet so you need to help me here.
Does the program use GA?
Why have you used the expression "good"? It immediately would make an adjudicator think that something bad has been sidestepped or hidden. Eg missing data.
I am not sure about missing data. You generate 10^5 pairs and then make four calculations apparently with about 55000 in each calculation. So you appear to have used replications, which was presumably a stipulation of Richard, and hence a slippery step not in your favour. And what happened to the other 45000 pairs of data generated?
It is a pity that the adjudicators gave no feedback. A checklist of "failure" points would have helped.

With best wishes.

[harry]

On this one I agree with Jaynes and Bell.

I'd welcome clarification here: What is being agreed to? Thanks.

Again:

Correlations can be nonlocal without anything weird going on, as Jaynes argued (and Bell admitted that in his socks paper).

"We can measure A in any direction we please; whereupon we can predict with certainty the value of B in the same direction. [..] Bell took it for granted that a conditional probability P(XjY) expresses a physical causal influence, exerted by Y on X.
[...but...] Fundamentally, consistency requires that conditional probabilities express logical inferences." (1989)
- http://neuroself.files.wordpress.com/20 ... teries.pdf

"But when you see [..] that the first sock is pink you can be already sure that the second sock will not be pink. Observation of the first, and experience of Bertlmann, gives immediate information about the second. There is no accounting for tastes, but apart from that there is no mystery here." (1980)
- http://cds.cern.ch/record/142461/files/198009299.pdf

[edit: As this is likely to "bug" others as it did me, the following remark may be useful. Note the slight misrepresentation of Bell by Jaynes: Bell admitted in 1980 that nonlocal (immediate) correlations could be just the result of logical inference. Thus he did not "take it for granted" (at least not anymore in 1980, maybe he did in 1964) but he made a very plausible sounding argument for physical influences further on in his socks paper of 1980.]

Once more, this whole debate will disappear like snow for the sun with the first simulation program (maybe based on the articles of this thread?) that is successful according to an independent expert panel. Who takes up the challenge?

[FrediFizzx]

Fred says that Richard wants to use flatland correlations in the simulation. (You may have to correct me here ... )

Ben, to be more exact, Gill wants to use flatland algebra to calculate the correlations in the simulation. Of course if you do that then you have "rigged" it so that no simulation will ever work to show strong correlations. Totally bogus way of doing the simulation if you can't simulate what we think Nature is doing in the first place. In the real experiment, no such simulation is needed so the correlations are just calcuated the standard way on the results. That will show that Joy is either right or wrong.

So to get back on topic, we have two recent classical experiments that show strong correlations. What is the explanation? The authors just relegate it to that they are simulating QM and that they have non-local "entanglement". Or we can explain it using Joy's classical local realistic model that explains the experiments and QM.

[Ben6993]

Hi Fred

That is complicated. The real experiment uses ordinary algebra but you say that the simulation needs the use of geometric algebra? Shouldn't the GA be used on hidden variables [zillions of them] of the ball fragments? I thought that stage was being bypassed [not surprisingly!] in the calculations. The vectors in the microscopic case are not observables in the real experiment, but they are observables in the macroscopic real experiment as they will be written into two files of vectors. In the microscopic case, the observables only exist after applying as and bs detector angles to the data files to obtain As and Bs. In the macroscopic case the raw data of vectors are observables and so should be treated with ordinary algebra in both real experiment and simulation. Isn't the root of the problem here, i.e. about using ordinary algebra or GA, the difference between micro and macro rather than between real and simulated?

I wouldn't say the there is anything rigged or bogus about the tests. It is just that they are designed by someone who sincerely holds a very different point of view. Time will tell who is correct.

Best wishes

[FrediFizzx]

Hi Fred

That is complicated. The real experiment uses ordinary algebra but you say that the simulation needs the use of geometric algebra? Shouldn't the GA be used on hidden variables [zillions of them] of the ball fragments? I thought that stage was being bypassed [not surprisingly!] in the calculations. The vectors in the microscopic case are not observables in the real experiment, but they are observables in the macroscopic real experiment as they will be written into two files of vectors. In the microscopic case, the observables only exist after applying as and bs detector angles to the data files to obtain As and Bs. In the macroscopic case the raw data of vectors are observables and so should be treated with ordinary algebra in both real experiment and simulation. Isn't the root of the problem here, i.e. about using ordinary algebra or GA, the difference between micro and macro rather than between real and simulated?

I wouldn't say the there is anything rigged or bogus about the tests. It is just that they are designed by someone who sincerely holds a very different point of view. Time will tell who is correct.

Oh for heaven's sake! It is not complicated at all and it has nothing to do with a particular point of view. This is all pretty standard scientific procedure. In the real experiment, Nature does what it does. The 3-sphere geometry is automatically done by Nature in the real experiment if Joy is right. In the simulation, you have to use some process to simulate what Nature is doing in regards to the 3-sphere geometry. There is nothing more to it than that. If you are reading more into it than that, then I can't help you understand it. You are stuck in flatland just like Gill.

[Ben6993]

Fred, As I said a long time ago, I support Joy in his microscopic model but find it hard to believe in the macroscopic model. I think that particles exist in many dimensions which is far from a flatland point of view. I'm off.

[FrediFizzx]

Then how do you explain the two recent classical macroscopic experiments that show strong correlations that this thread is about?

But besides that, it is clear to see that if Gill's flatland algebra process is used in any EPRB type simulation, that simulation will NEVER be able to show strong correlations no matter whether it is about the microscopic or macroscopic realms. Therefore, that process is "rigged" and indeed bogus. The bottom line is that it is not a fair evaluation of whether or not Joy's experiment will show strong correlation. There is only one way to find out. Do the experiment. Even if you don't think it will be successful, we can't be sure until it is done.

[FrediFizzx]

Once more, this whole debate will disappear like snow for the sun with the first simulation program (maybe based on the articles of this thread?) that is successful according to an independent expert panel. Who takes up the challenge?

LOL! That is awefully naive of you to think you would ever get any "experts" to go up against the Bell mafia. It's not going to happen.

IOW, you are going to have to decide for yourself and not rely on anyone else.

And... why the heck would you need a simulation if you have real experiments that show that Bell was wrong?

[harry]

[..] why the heck would you need a simulation if you have real experiments that show that Bell was wrong?

Once more, experiments cannot show Bell right or wrong - his theorem (the physical one) is a claim about theories.
However, those experiments show that the predictions are not messed up - and the prededictions are claimed to be "classical".
Normally this implies that they were done by means of "local realistic" models.

[FrediFizzx]

[..] why the heck would you need a simulation if you have real experiments that show that Bell was wrong?

Once more, experiments cannot show Bell right or wrong - his theorem (the physical one) is a claim about theories.
However, those experiments show that the predictions are not messed up - and the prededictions are claimed to be "classical".
Normally this implies that they were done by means of "local realistic" models.

Your logic here escapes me. I must be missing something. ???

[menoma]

Crowd-sourcing for Dr Christian's proposed experiment appears to be going nowhere. On the basis of that observation let us assume that the experiment is never going to be conducted absent some dramatic and undeniable failure of the "Bell Mafia" world-view. So what form might that failure take? If the quantum computation project implodes it might be explained by the practical difficulty of scaling up -- an engineering, not a theoretical catastrophe. Joyistas must look elsewhere for their In Hoc Signo Vinces. Where might that elsewhere lie?