Thoughts about Bell, Bohm, Christian, et al.

[minkwe]

Since others have derived it before me, I will just point you to the original CH74 paper, and the first pages of Christensen et.al. There are more authoritative sources than Wikipedia, you see.

Are you afraid to learn the conceptual errors in the derivation? Are you confident that the derivation is correct, and applies to epr-simple? If you are, then what are you afraid of. Go ahead and derive it step by step, laying down all the assumptions required as you go along. Trust me, you will learn a thing or two in the process, unless your mind is made up and you don't want to be "confused" by "strange" facts too.

[Heinera]

Since others have derived it before me, I will just point you to the original CH74 paper, and the first pages of Christensen et.al. There are more authoritative sources than Wikipedia, you see.

Are you afraid to learn the conceptual errors in the derivation? Are you confident that the derivation is correct, and applies to epr-simple? If you are, then what are you afraid of. Go ahead and derive it step by step, laying down all the assumptions required as you go along. Trust me, you will learn a thing or two in the process, unless your mind is made up and you don't want to be "confused" by "strange" facts too.

I have better things to do with my time than spoonfeeding you, when there are perfectly adequate derivations out there. Now, could it be that you have actually computed the lhs value of the inequality with your simulation, since you are so grumpy?

[minkwe]

It is the difference that if there are such detection failures, we have the detector efficiency loophole open, thus, the experiment does not allow to rule out local realism. And the simulation does not present a counterexample to Bell's theorem. In above cases, the experiment as well as the simulation do not reach their aim.

Summary: You do not see any difference in how the expectation values would be calculated with or without the zero outcomes, but you continue to claim that the zero outcomes are rejected in calculating the expectation values, and that makes the Again, simply repeating garbage that you've heard from else where without understanding exactly what it is all about. The only valid thing you have said is that Bell failed to include zero outcomes in his derivation. Yet for some reason, it is the simulation that is "bad" and "invalid".

Why should I care what boggles your mind? The sum is over a different set of outcomes, one symmetric to permutations between a,b,c,d, the other one asymmetric. We need symmtry in the proof of the theorem, and it is easy to construct counterexamples with asymmetric sets.

Are you using the word "symmetry"/"asymmetric" because you can't bring yourself to say counterfactual? Is the actual experiment symmetric or asymmetric? Is it possible to perform a symmetric experiment ever? Why-o-why do you think it is "invalid" to use zero outcomes in the simulation, but it is "valid" to compare the inequalities with "asymmetric" experiments?!!

Think think think: Your arguments are equivalent to
a) Bell did not use zero outcomes in the inequality, it is impossible to derive the inequality with zero outcomes, therefore experiments which use zero outcomes are invalid.
b) Bell relied on symmetry to proof the the theorem, it is impossible to proof the inequality without symmetry, therefore experiments and simulations which are asymmetric are invalid

Just change symmetry to "counterfactual outcomes" in statement b) and you will have the argument I have been making which you are trying hard not to accept. by making nonsensical statements like:

Bell uses the CFD in his proof.But the result is about averages, and these averages can be measured without measuring P(a,b), P(a,c) and P(b,c) on the same particles. So, it is the proof which uses CFD, but the BI[Bell inequalities] is not about them.

[Heinera]

The only valid thing you have said is that Bell failed to include zero outcomes in his derivation. Yet for some reason, it is the simulation that is "bad" and "invalid".

It is "bad", or "invalid", or just simply wrong to apply the original Bell's (as well as CHSH) inequality to your simulation, because Bell (as well as CHSH), in his/their proof assumes a simulation where all particles are detected. Now, how is your computation of the CH-inequality going? Found any violation there, so far?

[Schmelzer]

Summary: You do not see any difference in how the expectation values would be calculated with or without the zero outcomes, but you continue to claim that the zero outcomes are rejected in calculating the expectation values, and that makes the Again, simply repeating garbage that you've heard from else where without understanding exactly what it is all about.

Ok, let's explain this game for small children. I choose three cards, red or black. You can pick two of them, If they have the same color, you win. Do you have a chance to win? 1/3 is what Bell's inequality predicts. Because there are three cards, two colors, so at least two cards have the same color. But only if every game counts. If we introduce a zero, you will never win, because I choose one red, one black, and zero, if you choose the zero card, the game does not count, else you loose.

So, that the zeros allow to violate BI is a triviality.

The only valid thing you have said is that Bell failed to include zero outcomes in his derivation. Yet for some reason, it is the simulation that is "bad" and "invalid".

He has not "failed to include", but with zeros you cannot prove the theorem. Thus, experiments with zeros can not falsify local realism.

Are you using the word "symmetry"/"asymmetric" because you can't bring yourself to say counterfactual?

counterfactual, counterfectual, counterfictual, counterfoctual, counterf....

Why-o-why do you think it is "invalid" to use zero outcomes in the simulation,

Because it is well-known that a simulation using zeros can violate BI, thus, such a simulation is no more interesting than a simulation of 1+1=2.

Your arguments are equivalent to
a) Bell did not use zero outcomes in the inequality, it is impossible to derive the inequality with zero outcomes, therefore experiments which use zero outcomes are invalid.

Name them as you like, they are as interesting as the observation of an apple falling down, because they show as much new information about our world.

b) Bell relied on symmetry to proof the the theorem, it is impossible to proof the inequality without symmetry, therefore experiments and simulations which are asymmetric are invalid

The aim of introducing symmetry into the discussion was a small hope that it helps you to understand the point. A hint. It didn't, it has obviously confused you even more, so, forget about symmetry.

Just change symmetry to "counterfactual outcomes" in statement b) and you will have the argument I have been making which you are trying hard not to accept.

False statements are, of course, hard to accept. Because Bell's proof consists of two parts, the first of which, the EPR argument, you prefer to ignore, because it derives what he uses in the second part. And, then, you don't get the fact that Bell's inequality is about expectation values.

[minkwe]

The only valid thing you have said is that Bell failed to include zero outcomes in his derivation. Yet for some reason, it is the simulation that is "bad" and "invalid".

It is "bad", or "invalid", or just simply wrong to apply the original Bell's (as well as CHSH) inequality to your simulation, because Bell (as well as CHSH)

Bell required counterfactual outcomes in his simulation. Is it "bad" or "invalid" to apply the original Bell inequality to experiments which do not have counterfactual outcomes? Oh, you won't answer this, nor can you answer it because it would reveal you're clueless.

Now, how is your computation of the CH-inequality going? Found any violation there, so far?

What computation. You claim it applies to my simulation, I claim it does not. So it up to you to show that it applies by deriving the inequality, and posting the code you claim to have tested against the CH. What are you afraid of. Start deriving already, instead of putting your pants on fire with lies that you tested my simulation against the CH.

Where is the derivation of the CH.

[minkwe]

Ok, let's explain this game for small children. I choose three cards, red or black. You can pick two of them, If they have the same color, you win. Do you have a chance to win? 1/3 is what Bell's inequality predicts. Because there are three cards, two colors, so at least two cards have the same color. But only if every game counts. If we introduce a zero, you will never win, because I choose one red, one black, and zero, if you choose the zero card, the game does not count, else you loose.

So you prefer to go back to the contrived "children's" card game. We are talking about a pair of spin-1/2 particles that can only ever be detected once, and somehow you think bringing up cards is relevant? Which card among your three is counterfactual? Puhleese . epr-simple is not about cards, colors of cards. it is about the epectation value of paired outcomes calculated as .

The only relevant loophole in epr-simple is the fact that the set of particles used to calculate , is different from the set of particles used to calculate , etc. In other words, the only loophole is the fact that there are no counterfactual results being used in epr-simple. The claims about detection loophole are red-herrings. Presence or absence of 0 outcomes does not change the above expectation value, since there are no terms there which contain zero outcomes. However, the expectation value is affected by the presence or absense of counterfactuals, because the numbers of degrees of freedom are drastically different in both cases. And btw, this loophole, (lack of counterfactuals) is present in every CHSH-test experiment It is fatal. It cannot be remedied. My illustration a couple of pages ago was to show this fact, but you missed the boat and went completely astray because you chose to focus on an irrelevant "detection loophole".

He has not "failed to include", but with zeros you cannot prove the theorem. Thus, experiments with zeros can not falsify local realism.

He surely failed to include it because there is no physical principle which says all particles must be detected.

The aim of introducing symmetry into the discussion was a small hope that it helps you to understand the point. A hint. It didn't, it has obviously confused you even more, so, forget about symmetry.

I was just about to ask you to explain exactly what symmetry means. But I guess you don't even know as usual. Your position is inconsistent. The reason you give for why epr-simple is "invalid" should make all claimed experimental tests of Bells inequality invalid. It implies no test of Bell's inequality can ever be valid. Not even one with colored cards. Let me know when you find an experiment which produces counterfactual results.

Oh, I forgot, CFD is "derived", ..., from EPR , ..., and does not exist in QM, ..., except just a little, but it really is "predetermined" which is completely absent from QM, except in special cases, ..., but Bell's theorem proves that CFD is false, but that doesn't affect precious QM, QM is always true.., precious

False statements are, of course, hard to accept.

False statements are easily rejected. What is hard is when your mind is made up and you've invested your live chasing a viewpoint that is completely bogus and you don't want facts to get in the way of you "protecting your investment".

Because Bell's proof consists of two parts, the first of which, the EPR argument, you prefer to ignore, because it derives what he uses in the second part. And, then, you don't get the fact that Bell's inequality is about expectation values.

Duh, the EPR argument was made by EPR not Bell. Bell does not make any EPR argument.
Duh, we are discussing Bell's paper, not the EPR paper. I fully accept the EPR argument that QM is incomplete, and it is unreasonable talk about the revelation of Alice's result "creating Bob's eigenstate" as somebody mentioned not long ago. You are the one claiming on the basis of Bell's theorem that realism is non-local. Yet you've been unable to defend it. In fact, you now know how shaky that claim is, although I doubt you'll ever admit it.

[minkwe]

I have better things to do with my time than spoonfeeding you, when there are perfectly adequate derivations out there.

How convenient, you don't provide a derivation because you afraid you wont be able to defend it in the face of scutiny.

Now, could it be that you have actually computed the lhs value of the inequality with your simulation

Keep dreaming, epr-simple is a CHSH-test simulation not a CH one. Where's the code you claim to have tested it against the CH? Could it be your claim was a lie?

[minkwe]

If Bell forgets to consider hidden instrument parameters in his derivation, then instruments are forbidden from having hidden parameters.

They are not forbidden, the space of possible values of is arbitrary, thus, can contain instrument hidden parameters too. What matters is that there is predetermination, derived from the EPR criterion, which does not allow them to depend on a and b.

Let us see, is the outcome at Alice allowed to depend on hidden instrument parameters at Bob and vice versa? Please split up lambda into source and instrument parameters and you'll quickly have a realization. Let , now do the same to flesh out Bell's equations (1) and (2)
Bell's equation (1) becomes:
,

I leave equation (2) as an exercise. The integral only works if . Therefore either no instrument parameters are allowed or ALice's outcome must depend on Bob's remote instrument parameters for the derivation to proceed.

[Schmelzer]

... What matters is that there is predetermination, derived from the EPR criterion, which does not allow them to depend on a and b.

Let us see, is the outcome at Alice allowed to depend on hidden instrument parameters at Bob and vice versa? Please split up lambda into source and instrument parameters and you'll quickly have a realization. Let , now do the same to flesh out Bell's equations (1) and (2)
Bell's equation (1) becomes:
,

I leave equation (2) as an exercise. The integral only works if . Therefore either no instrument parameters are allowed or ALice's outcome must depend on Bob's remote instrument parameters for the derivation to proceed.

Of course, the space is arbitrary and can be splitted into . The points is that it is not allowed that depends even on a, it should be independent of a as well as b, it should be predefined, that means, defined before the choice is made what a, b to measure. And, of course, it is not allowed that the probability distribution becomes influenced by a using some .

The device itself exists, of course, a long time before the actual measurements are done, and their construction parameters can even allowed to be known at the other side, no problem at all. Then, the "must depend" is clearly nonsensical. One can allow that Alice's outcome depends on predefined device parameters of Bobs device without causing problems for the further derivation - it works in the same way.

[minkwe]

Of course, the space is arbitrary and can be splitted into .
...
it should be predefined, that means, defined before the choice is made what a, b to measure.

You like unicorn's don't you. I suggest you use your imagination for more than unicorns. Don't you see that in Bell's proof it is not allowed for instrument parameters, if they exist, to be dynamic? They are not allowed to change after the settings have been made but before the particles have been detected. In short, the instruments are static, all their atomic and subatomic particles must freeze as soon as Ilja decides to do a measurement. Otherwise please explain how you end up with the same on either side if can contain hidden instrument properties that are dynamically changing locally on each side. Do you have a hidden non-local tunnel hidden behind the unicorns? Or maybe Bell secretely assumed the superdeterminism of the instrument parameters.

The device itself exists, of course, a long time before the actual measurements are done, and their construction parameters can even allowed to be known at the other side, no problem at all. Then, the "must depend" is clearly nonsensical.

Left brain fighting the right brain again I see, notice how you say argue against "must depend" right after you argued it was okay for it to depend because the instrument existed long before the experiment so their parameters could have been known at the other side. Utter garbage.

Think. According to Bell, is not allowed. Your claim that can be arbitrary, is nonsense once you think about it for a second. Can only contain parameters that are known in the common light-cone of Alice and Bob. Your attempt to make instrument parameters fit by suggesting that they are either static and known in advance, or superdeterministic is nonsensical. And don't even get me started on the time issue, we've barely scratched the surface of it.

My point remains, with instrument parameters, Bell's integral only works if . Clearly you think this is "no-problem at all".

[minkwe]

If is allowed to be dynamic and time-dependent, then you have


--> DEADEND

To fix this you must make an additional assumption that , ie that both particles must be measured exactly simultaneously!!! No error in path-length of flight from source to station is allowed, and no time-delays of detection at each station is allowed. must be exactly the same as otherwise you can't proceed.

Some might be tempted to suggest that dynamic hidden instrument parameters are "no problem at all" because they could simply be absorbed into dynamic time dependent functions A and B. But as we've seen above, it is a dead end, unless you make an additional assumption which is impossible to achieve in practice, ie a global time of measurement. You end up with:



-> DEADEND

[Schmelzer]

Don't you see that in Bell's proof it is not allowed for instrument parameters, if they exist, to be dynamic? They are not allowed to change after the settings have been made but before the particles have been detected.

They are not allowed to magically disappear. They are not allowed to know in advance which direction will be measured. This is what matters. If they change, following some deterministic equation, nobody cares.

The point is that they predetermine the values. So, take a value of time before the choice of a and b is made. Then, the at that time should define the result A measured later, by a function , where the choice of a will be made only at and the observation yet later at . This is what predetermined means.

All you have to care is that they cannot be used for hidden information transfer. Say, let's consider the evolution of the instrument parameters at A after a has been chosen - they will, after this, somehow depend on a too. Then, of course, you have to care that these paramaters will not be used to compute B. But as long as you don't try such dirty tricks, (ups, sorry, invisible pink tricks) there is no problem with incorporating instrument parameters and even their evolution.

Otherwise please explain how you end up with the same on either side if can contain hidden instrument properties that are dynamically changing locally on each side.

The same or different on each side is, again, nothing which matters. As long as the parameters themself as well as their probability distribution do not depend on a, b. A mathematical triviality. If A depends on and B depends on , then I can define a big space containing them all, and have the same on above sides.

Think. According to Bell, is not allowed. Your claim that can be arbitrary, is nonsense once you think about it for a second.

There is no restriction on the choice of the space made in the paper. All one needs are two functions and a probability distribution on this space which does not depend on a, b. As long as instrument parameters fit into this scheme, the proof works.

Can only contain parameters that are known in the common light-cone of Alice and Bob. Your attempt to make instrument parameters fit by suggesting that they are either static and known in advance, or superdeterministic is nonsensical.

I don't try to make them fit, I simply inform you about the fact that the space is arbitrary, and that there is no such restriction in the text, because no such restriction is necessary for the proof.

My point remains, with instrument parameters, Bell's integral only works if . Clearly you think this is "no-problem at all".

You have guessed it, there is no problem at all. To introduce a simple split into the space does not help you, as long as you don't want to hide the choice a among the to hide a violation of locality.

And, of course, once the theorem works even for , it works as well if one, to avoid such dirty tricks, restricts it to

[minkwe]

They are not allowed to magically disappear. They are not allowed to know in advance which direction will be measured. This is what matters. If they change, following some deterministic equation, nobody cares.

So, take a value of time before the choice of a and b is made. Then, the at that time should define the result A measured later, by a function , where the choice of a will be made only at and the observation yet later at . This is what predetermined means.

That is not the only thing that matters. That they shouldn't know in advance the direction, is completely uncontroversial. The point is that you think that's all that matters. But it is not. You haven't been paying attention to the argument:

Do you understand why Bell's inequality can be derived from


But cannot be derived from


If you do not yet understand this, maybe your puppet master can explain it to you.
Please think for a just a moment. Do you believe , if ?

If not then please look at the second equation on Page 406 of Bell's proof and tell me how you are going to factor out the A(...) function from an expression which contains two different A functions measured at a different time? You can't have it both ways. Either the functions are the same and the times are different, or if you want to force the times to all be t_0, then you are hiding all the differences in the functions.

Do you believe

The point is that they predetermine the values.

The functions do predetermine the values but are not necessarily the same so you need to forget about predetermination for a moment and think about what I'm telling you. This "predetermination idea" is blocking your thinking juices. Bell's derivation does not work if the are different in any way whatsoever. It doesn't matter whether you have different functions acting on a fixed , or fixed functions acting on different . As long as you have dynamics, you must assume that the everything is measured at exactly the same time. Otherwise no factorization, and therefore no inequality.

The same or different on each side is, again, nothing which matters. As long as the parameters themself as well as their probability distribution do not depend on a, b. A mathematical triviality.

Wrong. It matters. If is time dependent, then the probability distribution of is also time dependent, and then the if the measurement times of of A_a, A_b, A_c are all different from each other, the outcomes will correspond to different probability distributions of , and your so-called "mathematical triviality" becomes physical nonsense. For time dependent hidden variables, you must measure at exactly the same time to have the same distribution of .

There is no restriction on the choice of the space made in the paper.

There is a restriction that the probability distribution of must be the same for all terms in the inequality. Therefore the inequality cannot be derived for any for which the distributions in P(a,b) is different from those in P(a,c), such as for time-dependent s.

All one needs are two functions and a probability distribution on this space which does not depend on a, b. As long as instrument parameters fit into this scheme, the proof works.

Wrong. may not depend on (a,b), but may depend on , then if the two functions also depend on , , then even though everything is local, you will still have the outcomes varying with the settings, because different settings are active at different times. You must make the additional assumption that the measurements at all settings are all done at the same time. ie, fixed time. Otherwise the derivation fails.

[FrediFizzx]

It is quite amazing that after Michel has presented an extraordinary amount of incontrovertible evidence that the Bell diehards fail to accept the truth.

[Schmelzer]

The point is that you think that's all that matters. But it is not. You haven't been paying attention to the argument:
Do you understand why Bell's inequality can be derived from

But cannot be derived from

Of course, that's a triviality.

Please think for a just a moment. Do you believe , if ?

Why should I believe such things?

If not then please look at the second equation on Page 406 of Bell's proof and tell me how you are going to factor out the A(...) function from an expression which contains two different A functions measured at a different time?

Why should I even try to care? All I have to care about is the case considered in Bell's paper.

You can't have it both ways. Either the functions are the same and the times are different, or if you want to force the times to all be t_0, then you are hiding all the differences in the functions.

I couldn't care less. Choose whatevery time you like, using whatever notion of contemporaneity you like, name it , all one needs that it has to be before the events when a and b are chosen. Then, choose the which predefines the A,B, and the particular functions which describe how these particular values predetermine A and B. And you can continue with the proof.

For different you obtain different parameters and different functions? So what, if all one needs to prove the theorem is that one has some such functions? Use the most beatiful moment, the invisible pink one.

Wrong. It matters. If is time dependent, then the probability distribution of is also time dependent, and then the if the measurement times of of A_a, A_b, A_c are all different from each other, the outcomes will correspond to different probability distributions of , and your so-called "mathematical triviality" becomes physical nonsense. For time dependent hidden variables, you must measure at exactly the same time to have the same distribution of .

Of your, if you do nonsensical things, the result is nonsense. Why should we care about this? We would care if Bell would have done nonsensical things, he didn't. He has proven that the results are predetermined, thus, can be computed from knowing the of some much earlier time, earlier than all the measurements. This is sufficient to prove the theorem. Sufficient even if all the things which happen after happen at various different times, without a problem.

You propose some variant where this no longer works? A really big achievement.

All one needs are two functions and a probability distribution on this space which does not depend on a, b. As long as instrument parameters fit into this scheme, the proof works.

Wrong. may not depend on (a,b), but may depend on , then if the two functions also depend on , , then even though everything is local, you will still have the outcomes varying with the settings, because different settings are active at different times. You must make the additional assumption that the measurements at all settings are all done at the same time. ie, fixed time. Otherwise the derivation fails.

You don't get the simple point that as the time I can take any before the actual choices are made and observations done. This gives functions and a probability distribution . This can be done because it has been proven that the values are predetermined. All what matters in relation with time is that at the time they already have been predetermined, thus, to use the Big Bang time would not be appropriate, instead, one would have to use so that the past light cones of A and B no longer have an intersection at , so that the EPR argument works at and one can prove predetermination at .

[minkwe]

So finally you see the problem -- that's progress. But you don't care? Why should you?

You should care because you write rubbish like:

Bell uses the CFD in his proof.But the result is about averages, and these averages can be measured without measuring P(a,b), P(a,c) and P(b,c) on the same particles. So, it is the proof which uses CFD, but the BI[Bell inequalities] is not about them

You should care because you vehemently attack arguments like:

Bell does not use 0 outcomes in his proof.But the result is about averages, and these averages P(a,b), P(a,c) and P(b,c) can be measured even if 0 outcomes are present. So, it is the proof which uses only nonzero outcomes, the BI[Bell inequalities] is not about them

Claiming loophole, loophole. Failing to see that you are applying a double standard to simulations like epr-simple on the one hand, and QM and experimental tests on the other. Either they are all valid or all invalid. Either way, Bell's theorem is toast. If they are all valid, then ep-simple is an effective counterexample. If they are all invalid then Bell's theorem is false, and Bell's inequalities are irrelevant to physics.

You should care because Bell's inequality can only be derived if the averages P(a,b), P(a,c) and P(b,c) are measured at exactly the same time on exactly the same set of particle pairs. But this is impossible. You see, what is trivial to mathematicians can be very serious issues physically. That is why mathematicians without sound physical intuition should not be developing physical theories. They would simply produce junk. You should care about that.

You should also care because you go around Internet fora propagating falsehood that Bell's theorem forbids local hidden variable theories. Now you know that it does not, because the assumptions required can never be fulfilled in any humanly possible experiment. Now you know that the QM expectation values refer to three disjoint sets of particle pairs measured at different times, while the Bell inequality requires that the averages be done on exactly the same set of particles, at the same time. So Bell's theorem is simply false.

You should care because your ether theory is developed with the false understanding that realism is nonlocal. But you now know that is false, since Bell's inequality should never be compared to QM expectation values or experimental averages, since it does not apply to cases where the outcomes are from disjoint sets of particle pairs measured at different times.

Finally, you should care because:

the Bell proof, if you actually come to grips with it, falls apart in your hands! There is nothing to it. It’s not just flawed, it’s silly. If you look at the assumptions it made, it does not hold up for a moment. It’s the work of a mathematician, and he makes assumptions that have a mathematical symmetry to them. When you translate them into terms of physical disposition, they’re nonsense. You may quote me on that: the proof of Bell is not merely false but foolish.

Etc, etc. Just because there are many reasons why you should care, doesn't mean you will care. Many people live their lives in denial. You will have plenty of company who are "careless".

[minkwe]

You propose some variant where this no longer works? A really big achievement.

Yes! I propose a variant which matches more closely to the physical situation than the one Bell chose, and despite your claims that it is "no problem at all", I show that it does not work as Bell and you think it does. That you finally admit it does not work is a big acheivement.

You don't get the simple point that as the time I can take any before the actual choices are made and observations done. This gives functions and a probability distribution .

I get the point that the derivation only works if time is fixed so that all the terms correspond to exactly the same time. But you don't get the point that this is what your argument implies. You don't get the point that I already told you you could make the derivation work with fixed time. You don't get the point that such trivial mathematical "duct tape" becomes nonsense once you start thinking about the physics of actual experiments.

We would care if Bell would have done nonsensical things, he didn't. He has proven that the results are predetermined, thus, can be computed from knowing the of some much earlier time, earlier than all the measurements.

Oh somebody forgot to tell experimenters not to do nonsensical things like measure different averages at arbitrary times. They should be calculating the averages using the outcomes from a much earlier time than the measurements.

[Schmelzer]

So finally you see the problem -- that's progress. But you don't care? Why should you?

Learn to read. If I would see a problem, I would care. But if I see only meaningless, hm, [self-censored] modifications and questions, why should I care?

Claiming loophole, loophole. Failing to see that you are applying a double standard to simulations like epr-simple on the one hand, and QM and experimental tests on the other. Either they are all valid or all invalid. Either way, Bell's theorem is toast. If they are all valid, then ep-simple is an effective counterexample. If they are all invalid then Bell's theorem is false, and Bell's inequalities are irrelevant to physics.

Complete nonsense. Bell starts with theoretical assumptions, in particular Einstein causality, and derives an inequality. This inequality is, in principle, open to experimental tests. The ideal test would allow to falsify the assumption of Einstein causality (in the stronger meaning of "causality" supported by Einstein himself).

But the real experiments have, of course, limitations, which have to be considered. A simple one is the detector efficiency loophole, but there are other ones. Up to now, only the detector efficiency loophole has played a role here, simply because all your simulations use it. There is no double standard, simulations which use detector efficiency loopholes are rejected as invalid refutations, and experimental tests with detector efficiency loophole are rejected as experimental refutations of Einstein causality.

But it seems you don't understand the key point how Bell circumvents the necessity for having all outcomes out of the same experiment. It is part of the formula



The probability density and the functions of this formula describes, in a realistic theory, the state which results from a given preparation procedure. That means, if you use the same preparation procedure, the realistic theory predicts also the same P(a,b). This is the physical meaning of this equation. And that's why you can use the same preparation procedure to measure, in different rounds, P(a,b) for different choices of a and b.

And the simulation has to follow the same scheme - it has to define the functions and the random set which approximates some fixed probability distribution in the preparational part, without any contact with actual values a,b. And only after this a- and b-independent preparation one can compute the P(a,b), P(a,c), P(b,c).

You have noted that if one uses exactly the same sets to compute them, the BI will hold exactly. Instead, if one uses different sets, minor violations are possible, which corresponds to the fact that different subsets of values of can all approximate the same fixed probability distribution , but not exactly, but only approximately, while the BI are derived assuming exactly. But these are unsystematic rounding errors, and not relevant, with large enough numbers of values they decrease.

The situation is completely different if you choose different to compute P(a,b), P(a,c), resp. P(b,c), or, in other words, some . In this case, there can appear arbitrary large violations of the BI which don't go away with increasing numbers.

You should care because Bell's inequality can only be derived if the averages P(a,b), P(a,c) and P(b,c) are measured at exactly the same time on exactly the same set of particle pairs.

Which is simply wrong, Bell has derived it in a different way, using the EPR argument to prove predetermination.

the Bell proof, if you actually come to grips with it, falls apart in your hands! There is nothing to it. It’s not just flawed, it’s silly. If you look at the assumptions it made, it does not hold up for a moment. It’s the work of a mathematician, and he makes assumptions that have a mathematical symmetry to them. When you translate them into terms of physical disposition, they’re nonsense. You may quote me on that: the proof of Bell is not merely false but foolish.

Ok, I quote you with this, and the only point of quoting it is to show that what you claim is not merely false but foolish. Which is sufficiently self-evident now for the reader.

[minkwe]

So finally you see the problem -- that's progress. But you don't care? Why should you?

Learn to read. If I would see a problem, I would care. But if I see only meaningless, hm, [self-censored] modifications and questions, why should I care?

I've made my point clearly enough, even a caveman should be able to understand it. I said it before, it is not possible to have an insightful discussion with you. Everyone else reading this who wants to understand the issues, has understood them. So feel free to continue being "careless".

Complete nonsense. Bell starts with theoretical assumptions

Indeed Bell's assumptions are completely nonsense when you translate them to physics.

But the real experiments have, of course, limitations, which have to be considered. A simple one is the detector efficiency loophole, but there are other ones. Up to now, only the detector efficiency loophole has played a role here, simply because all your simulations use it. There is no double standard,

Hehehe All my simulations use a single loophole -- the counterfactual loophole, ie the absence of counterfactual correlations, which is present (and will be present) in all perform-able experiments. I encourage you to study both epr-simple, and epr-clocked before you make nonsensical claims about them. 100% of emitted particles are detected in epr-clocked. There is no detection loophole there. I just proved to you that all the loopholes are actually flaws in Bell's derivation and due to the fact that his assumptions are nonsense when you translate them to physics. The predictions of QM have been verified experimentally for years now, yet it has been more than 50 years since Bell's derivation, and yet nobody has found a way to do a so-called "loophole-free" experiment. And the fact that they keep trying to do one tells you they are as clueless as you are about Bell's blunder. I have just proved to you that there is no such thing as a "loophole-free" test of Bell's theorem, because the derivation itself makes it impossible due to it's nonsensical assumptions. There is a double standard, whether you like to admit it or not.

1) Bell's inequality cannot be derived if 0 outcomes are allowed.
2) Bell's inequality cannot be derived unless counterfactual outcomes are allowed.
3) Bell's inequality cannot be derived unless all outcomes are measured at the same time.
4) Therefore, Bell's inequalities do not apply to averages from experiments/simulations in which 0 outcomes are present.
5) Therefore, Bell's inequalities do not apply to averages from experiments/simulations in which counterfactual outcomes are not absent
6) Therefore, Bell's inequalities do not apply to averages from experiments/simulations in which are measured at different times

You want to argue (1 & 4) while at the same time refusing to see that (2 & 5) and (3 & 6) follow from the same logic. Simultaneously believing two contradictory arguments is called dishonesty if done with understanding, and stupidity if done without understanding. Which one is it?

If you don't yet understand this, you will never understand it. You prefer to repeat completely debunked canned phrases you've practiced regurgitating time and time again. Good luck. I'm done here. You can go back and read the last 15 pages of this thread to find detailed debunking of every claim you can possibly make in the future but I have better things to do than convince a blind man that the sun exists even though he can't see it.

simulations which use detector efficiency loopholes are rejected as invalid refutations, and experimental tests with detector efficiency loophole are rejected as experimental refutations of Einstein causality.

Experiments which do not have counterfactual outcomes must also be rejected as invalid refutations of Einstein causality. Experiments which do not measure all the expectation values at the same times must also be rejected as invalid refutations of Einstein causality. And since no experiment with counterfactual outcomes is possible, and no experiment in which all the expectation values are measured at the same times is possible, Bell's inequalities are irrelevant to physics.

Deal with it! It is probably time to go back and read Schulz's response to your comment on his paper. If you've learnt anything the last few days, you may begin to understand what he was trying to tell you back then. If not, then there is no hope left for you.

Continue spreading falsehood about how Bell's theorem rules out Einstein local causality. Cement your place in history as one of agents of the dark-age of theoretical physics. You are doing a good job of it already, and you have excellent company in a number of mathematical statisticians-turned theoretical physicists.

BTW, anyone who is clueless enough to think time is no problem at all, should try to derive Bell's inequality (following Bell's derivation from page 406) starting from:



Checkmate! I'm done here. This will be my last post on this issue. See ya! Auf Wiedersehen! Arrivederci!