On Tue, Aug 6, 2019 at 11:38 PM Bruno Marchal <[email protected]> wrote:
> On 6 Aug 2019, at 14:46, Bruce Kellett <[email protected]> wrote: > > > The problem here is that you are confusing the operation of the QC while > it is doing the calculation with the operation of reading the output. > > Because I study the QM applied to the couple “observer + observed”. > The operation of the quantum computer is not "observed". It is only the result that is observed. Then you can apply QM to the observer, but if the algorithm of the QC is effective, only one result is output -- there is no superposition of different possibilities. If there were, your QC would not be reliable, giving the wrong result with a non-zero probability. The operation is completely independent of any observer and any brain > state; it corresponds to unitary rotations of the state vector. > > Yes, in my case too. > > Reading the output corresponds to projecting this vector on to an output > set of basis states. > > In the relative way, or you are bringing some collapse of the wave in the > picture. > Se above. If the output qbits are in a superposition, your QC is not reliable. If the QC does its task effectively, the output basis qbits will be put > into definite states, > > Relatively to the observer, but in the global state, the observer will > inherit the superposition state, by linearity of the tensor products and of > the evolution. > As I said to Saibal, why are you guys so keen on retaining the superposition when it has no possible practical significance? Particularly in the case of the output of the QC, when a superposition of results means an ineffective computer. so the probability of the desired result approaches unity, and there are > effectively no "other Everettian worlds" in which different results would > be obtained. > > > > Then some branch of the superposition disappeared, and QM do no more apply > to the couple “observer-observed”, and we are out of Everett’s Relative > state theory. > Not so. This is precisely what Everett predicts for a quantum eigenstate. > Note that I don’t really believe that the word “world” makes sense. I > prefer to use “history” instead. Eventually, those will be computations, > and they will not need any world at all (just a tiny segment of the > arithmetical truth). > > All your talk about superpositions of observers is just a distraction from > the main issue, which you seem to accept (i.e., the basic vector analysis), > but the consequences of which you do not wish to acknowledge. There are no > "parallel worlds" in which the computations take place. > > A projection is not unitary. It does not represent some evolution > “out-there”. The observation, as you seem to describe it, is not obeying to > a quantum (unitary) transformation. > For the quantum computer, of course it is unitary -- or else the QC would not be reliable. ..... > There is no serious base problem, once we understand that the observer has > chosen the base corresponding to sharp measurement result that he can > memorise. > So the observer chooses the basis -- it is not part of the Schroedinger evolution? People who believe in quantum mechanics need to do better than this. > Whatever base is used, the relative state theory gives the sema results > for the measurement made in a base in which we can describe the (mechanist) > working of his brain and sensory apparatus. > In other words, you have to choose the basis that is stable under environmental decoherence. Choosing a different basis will not give you the observed result. And obtaining agreement with observation is, after all, what physics is about. Even Everett, bless his cotton socks, knew this. > To pick out one set of base vectors and claim that these vectors represent >> a set of parallel worlds in which the computations actually occur, is >> simply unnecessary -- description in terms of the single summation vector >> eliminates this stupidity. >> >> >> Unfortunately, we cannot eliminate the fact that an observer looking a >> the cat will use some position base to get the memory of the result of its >> observation, and without collapse, *whatever base* you are using to >> describe the overall situation (including the cat and the observer and its >> memory) the observation of the cat will lead to an observer seeing only the >> cat alive, and one seeing only the cat dead, and I don’t see how you will >> make this superposition disappearing. >> > > Still irrelevant. > > > I don’t see why. > Of course your observations are irrelevant -- just an attempt to divert attention from the fact that you have no answer to my main points. > Maybe you can tell me what happens in that situation. Note that even after >>> measurement, we can get back the interference effect by erasing the >>> memorised outcome of the result. Without collapse pure state remains pure, >>> and decoherence is relative to each “copies” of the observer in the terms >>> of the (universal) >>> >> >> >> These observations are entirely beside the point. You cannot erase the >> memory of the result because memory is intrinsically irreversible. >> >> >> Quantum mechanics is reversible. >> > > Hmmmm! Measurement results are generally not reversible. Zurek talks of > the environment as witness, and quantum Darwinism -- the many copies of the > result in the environment are not reversible, even in principle. Can you > pursue the escaping IR photons and turn them round? > > In principle. And that is enough to address the interpretation of QM > problem. > No, in principle it is impossible. Can you move faster than light to catch the stray IR photons? All your talk of superpositions the persist "in principle", is just so much distracting hot air. The only person you are fooling is yourself. > Quantum erasure is a technical matter that occurs only in highly >> constrained situations. I think you should catch up on some recent work on >> quantum foundations, in which Everett does not necessarily require the >> continuing purity of the quantum state. Measurement changes the pure state >> into a mixture. Zureck has made considerable progress in this direction in >> recent years. Quantum foundations has moved on since 1957. >> >> >> Mixture are relative personal outcome for the observer in the >> superposition state. There are no mixture in the global universal wave >> function, or you are introducing a wave packet reduction somewhere. >> > > The global universal wave function is a non-operational entity. > > The question addressed here is conceptual. > Concepts have to conform to experience or they are empty and useless. The purpose of science, after all, is to understand our experience -- not to construct imaginary worlds that bear no relation to anything at all. > Measurement changes nothing, that is why Everett allows (and use) >> Mechanism. Bohr and wave-reduction philosophy require the observer to >> disobey quantum physics, and it uses a dualist (and unknown) theory of mind. >> > > All irrelevant to the issue at hand. > > I don’t see this at all. You are saying that the many terms of a > superposition becomes one term after an observation, which I say would > entail a theory where the observer is no more described by quantum > mechanics. > Then you clearly do not understand the quantum mechanics of the operation of an effective quantum computer. If the output result is not definite, the QC has failed to work according to specifications. Nothing can make a superposition into a pure state, except relatively to > the first person associated to the (mechanical) observer. Or some collapse > of the wave is at play, but that is not a unitary evolution. > So the observer does see a definite outcome, does he? Bruce -- You received this message because you are subscribed to the Google Groups "Everything List" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/everything-list/CAFxXSLSex6nmEW_5sdi_X-yXH8k6J2736tWR1ZxwM3ozH1pzTQ%40mail.gmail.com.
