"Simulating physics with computers" by Feynman, IJTP 21 1982
Reference "Feynman and Computation" ed. A. J. G. Hey
This is Vol II of Feynman's Lectures on Computation.
11.2 "Simulating Time ... space-time view, imagining that all the points of space and time are all laid out, so to speak ahead of time... a 'computer' rule ... it's not the standard kind of computer which operates in time... The state si at the space-time point i is a given function Fi(sj,sk...) of the state at the points j, k in some neighborhood of i ... if this particular function is such that the value of the function only involves the few points behind in time, earlier than this time i, all I've done is to redescribe the cellular automaton ... let's think of a more general kind of computer... a wider case ... If F depends on all the points both in the future and the past, what then? ....11.5 Can Quantum Systems be Probabilistically Simulated by a Classical Computer? ... The only difference between a probabilistic classical world and the equations of the quantum world is that somehow or other it appears as if the probabilities would have to go negative..."Turning to quantum mechanics.... secret, secret, close the doors! we always have had a great deal of difficulty in understanding the world view that quantum mechanics represents ... It has not yet become obvious to me that there's no real problem. I cannot define the real problem, therefore I suspect there's no real problem, but I'm not sure there's no real problem. So that's why I like to investigate things."
The evidence:
I. "let's think of a more general kind of computer... a wider case ... If F depends on all the points both in the future and the past, what then? ....
II. The only difference between a probabilistic classical world and the equations of the quantum world is that somehow or other it appears as if the probabilities would have to go negative..."
My conclusion, the negative "classical" probabilities required by specifically quantum phase coherent q-bit effects ("superposition principle", "nonlocality") are "backward causations".
That is, positive "classical" probabilities are retarded propagating from past to present. Negative "classical" probabilities are advanced propagating from future to present.
Reference: Huw Price "Time's Arrow and Archimedes Point" develops this in detail.
Thanks to Marcello Truzzi for pointing out November 27 New Scientist's "It's heresy, but time running backwards could explain dark matter."
1959 Feynman creates nanotechnology with "There'e Plenty of Room at the Bottom." Contrary to all expectations things get better as we scale down to nanotech. p. xii
Marvin Minsky says no to quantum uncertainty and yes to quantum certainty in the sense of Viki Weisskopf's "Knowledge and Wonder" MIT Press. Minsky seems to refute Stapp on free will from quantum uncertainty?
Benioff writes on "quantum robots" - disappointing paper but I will take another look eventually.
Carver Mead also a Cal Tech Professor likes the EM vector potential A! Feynman wrote:
"In the general theory of quantum electrodynamics, one takes the vector and scalar potentials as fundamental quantities in a set of equations that replace the Maxwell equations." p.21
In Vol II 15-4 of Feynman's 1963 Lectures in Physics to Cal Tech Sophomores - that's when I first met Feynman driving up from Ford Philco Aeronutronics, Newport Beach, CA in my black XK 150 Jaguar convertible with white leather and knock-off wire wheels. He liked the car! :-) Fred Alan Wolf was at Hughes then where Feynman also went once a week.
"is the vector potential a real field? ... a real field is a mathematical function we use for avoiding the idea of action at a distance ... a real field is .... a set of numbers we speciy in such a way that what happens at a point depends only on the numbers at that point."
This is a local real field. Bohm's quantum information field is a nonlocal real field of q-bits.
Feynman also says that the ambiguity in the 4-potential from the internal symmetry gauge transformations (in a given gauge constraint) is not relevant to the physical reality of it anymore than the ambiguity of the Maxwell field tensor under Lorentz space-time transformations. Note, time-dependent constraints do work on the system. Suppose we have a time-dependent gauge constraint? Would this mean that the 4-potentials can be engineered to directly do work under proper conditions? What about the Bohm-Aharonov shift of electron wave fringes?
"In our sense then, the A field is real." (Feynman cited by Carver Mead p. 22)
Carver Mead writes:
"For a real electric motor, the B field is concentrated in the iron, rather than in the copper wire where the current is flowing, and the equation [ i.e. F = q(E + v x B) ] gives the WRONG ANSWER (caps mine) BY A FACTOR OF MORE THAN 100 ! That factor is due to the failure of B to be real, precisely in Feynman's sense." p, 23
This remark gets to the heart of the matter for some of the anomalies we are seeing.
Remember the classical electromagnetic field 4-potentials directly control the phase of the quantum information field of the material charged source.
"In those days physics was an openly combative subject - the one who blinked first lost the argument. Bohr had won his debate with Einstein that way, and the entire field adopted the style. Feynman learned the game well - he never blinked. ... If Feynman was stuck ... he had a wonderful way of throwing up a smokescreen; we used to call it 'proof by intimidation'..." written by Feynman's Cal Tech colleague Carver Mead p. 25
On "cargo cult" numbo jumbo and the Laputan formalism fetishism of p-adic Matti, Larry, and Harry (who's Harry?) that we sometimes see on this forum and elsewhere especially in more respectable mainstream theoretical journals Feynman who "had no use for a theory devoid of physical content" (p.26) said:
"If there is something very slightly wrong in our definition of the theories, then the full mathematical rigor may convert these errors into ridiculous conclusions .... carrying rigor to the point of rigor mortis ... it is the facts that matter not the proofs ... if the facts are right the proofs are a matter of playing around with the algebra correctly."
"On the quantum information theory of C. H. Bennett of IBM.
Coherent transmission and transformation of "intact quantum states" really requires the Bohm ontology not Bohr's epistemology. Quantum states are nonlocal objectively real individual physical objects of nonclassical information in the Bohm ontology, not merely statistical data book keeping algorithms of internal representations of the universe inside sentient minds of Wheeler's "observer-participators".
Bennett's theory is restricted to non-sentient unitary time evolutions of the patterns of quantum information. There is a non-unitary "R-collapse" that Bohm eliminates. In Bohm's fundamentally non-probabilistic ontology, the R-collapse is simply the capture of the material Bohm point X(t) in an attractor basin on the quantum information landscape.
Shannon's maximal c-bits of equiprobable binary choices are replaced by qu-bits i.e. coherently superposed "two state" quantum systems. What is the real physical difference between the Shannon c-bit and the qu-bit? Feynman showed that the real difference is one of negative probability! That is, the specifically quantum weirdness can be described as negative probability! I mean if you want to use ideas from classical probability theory, that Bohm avoids altogether at the foundational ontic level, you can do so, but the price you must pay is negative probability! Quantum wave interference really means negative probability. The Pauli exclusion principle, Bose-Einstein condensation, chemical bonds, none of them would exist, we would not exist, without these weird negative classical joint probabilities that you see in the "wavelet" Wigner phase space density formalism for example. Feynman wrote:
"The only difference between a probabilistic classical world and the equations of the quantum world is that somehow or other it appears as if the probabilities would have to go negative..," p.145
So how can we understand this? We can understand it as "backward causation" in which the future literally reaches backward in time to co-create the present with the past. For more details on how backward causation explains this quantum weirdness of negative probabilities in which an electron apparently is in several places simultaneously see Huw Price's book, "Time's Arrow and Archimedes Point." For example, the caged electron in tubulin is a qu-bit. In fact, it may be The Eccles Gate where mind meets matter and consciousness is generated in the feedback of that electron to its quantum pilot wave that it shares with 10^18 other caged electrons all over the human brain to holographically (Pribram) generate a single conscious moment of about 0.3 seconds. Note the Penrose-Hameroff picture of this is very different (not holographic) using only about 10^11 electrons for each moment rather than the entire brain.
C. H. Bennett writes:
"Just as any transformation of classical data can be expressed as a sequence of simple gates (e.g. NOT and AND) acting on the bits one and two at a time, any transformation of quantum data can be expressed as an array of two-qubit controlled NOT-gates (... CNOT or XOR) and one-qubit unitary rotations. The central feature of quantum data processing is the superposition principle." p. 178
In more dramatic picturesque, but equally accurate language giving a new perspective, the central feature of quantum data processing is negative classical probabilities coming in to the present from the future i.e. advanced or backward causation.
Superposition in, superposition out. Gates are connected by "wires" that store and move qubits from gate to gate. The set of qu-bits in a string are generally "entangled" i.e. nonlocally connected forming a common pool of information. You cannot think of the entangled qubit as separated from its entangled partners.
Unentangled pair of qubits input to a CNOT gate has an entangled pair output. The CNOT gate nonlocally glues the q-bits together joins them at the hip like Siamese Twins. One qubit alone looks locally completely random, yet together one sees a coherent reproducible nonrandom nonlocal order. That is the two spatially separated random quantum processes are nonlocally synchronized in perfect harmony at a distance faster than the speeding photon. You cannot, however, use this effect to transmit a useful message faster than light. You can see the nonlocal synchronization in the backward light cone of your local detector.
Entanglement necessary, but not sufficient, for experience?
"A classical wire .... as a kind of noisy quantum channel, in which the incoming qubit interacts via a CNOT with an ancillary qubit, which is then dumped into the environment. . The wire's environment makes a nondemolition measurement in the |1>, |0> basis, on the data passing through." p. 178
"integer factoring can be sped up exponentially on a quantum computer" p. 179 This means goodbye to computer security based on the difficulty of factoring large "molecular" composite integers into their "atomic" primes. Each integer is a code, a finger print.
Let p(i) be the ith prime. Each integer is then the infinite sequence of integers {n(1), n(2), .....} of powers of the prime factors. Note that p(1) =2, p(2) = 3, p(3) = 5, p(4) = 7, p(5) = 11, p(6) = 13, p(7) = 17, p(8) = 23 .... So we have a denumerable infinity of primes. It is more interesting to look at archetypal patterns in finite fields of integers modulo primes. This would be a good project for a smart high school Whiz Kid - actually for a team of Whiz Kids.
"NP-type optimization and search problems, can be sped up quadratically, while yet other problems cannot be sped up at all." p. 179Only quantum computers can simulate general quantum systems. You can make classical computers simulate some quantum systems, but you can make a quantum system that no classical computer can simulate. p. 180
So far no practical quantum computing devices exist in 1999.
Efficient quantum error-correction codes only recently understood.
Decoherence per gate must be held down to at least 10^-6/2pi of relative phase between |1> and |0>.
Bennett is actually pessimistic comparing the likelihood of quantum computing to that for controlled hot fusion! p. 180
Is quantum computing a hype? Is it Much Ado About Nothing? Why is IBM spending so much money on it? These are rhetorical questions. The work in quantum computing is very important. However, mainstream physics uses a double standard when they call cold fusion and parapsychology "pseudoscience" in the face of their expensive failures like hot fusion.
Is this The Right Stuff To Make Star Trek Real? Ad Astra But first, onward to Cydonia on Mars. Let's see what is really there? If anything? Natural or artificial?