setAI said:
the only interpretation of QM that succesfully predicts quantum computing is the MWI- Deutsch has DEFINED quantum computation as computation across parallel universes-
Well, I hate to disturbe a fellow MWI-er, but I fail to see how one can say that quantum computation proves MWI. Honestly, quantum computation is nothing else but the unitary evolution of states of a system, and it is sufficient to put the Heisenberg cut AFTER obtaining the final result to be in agreement with Copenhagen, no ?
Imagine that the Heisenberg cut comes in for objects heavier than 200 tonnes. In other words, systems of more than 200 tonnes are classical systems and obey a superselection rule that only allows them to be in nearly classical states, and not in superpositions of these, and this is done by the projection postulate. How does quantum computing prove this statement wrong ? And if it does, if we now put the Heisenberg cut at 10^10 kg objects ?
the experimental evidence of quantum computation- specifically independant CNOT operations carried out in parallel on single photons- physically demonstrates the MWI- and demonstrates that non-multiverse interpretations are unphysical- according to the leaders of the field-
I fail to see how, for instance, Bohmian mechanics would give different observable results than MWI-QM. Now of course, I am also sensitive to the argument that Bohmian mechanics has some MWI flavor to it
Qunatum Mechanics IS the physics of parallel universes
Well, I also find this the most obvious interpretation of the formalism, but on purely formal grounds. I don't see how one can experimentally PROVE this view, given that it is empirically equivalent to Copenhagen with the Heisenberg cut far enough away, or to Bohm. It is for formal ugliness that I don't like these interpretations, not for their empirical falsifiation.
In fact, the experiment that "proves" to me MWI up to a point is an EPR experiment, but unfortunately, one needs another hypothesis for that, which was historically the first to be "tested" by these experiments.
The reason is that in order to get an indication of the validity of MWI, that one needs to put in superposition, and show interference, of VERY MACROSCOPIC systems (ideally, say, a planet).
Now, this is practically impossible to do because of environmental decoherence, which makes the interference invisible... unless one finds a way of making sure that the two branches that have to interfere cannot entangle with the same environment (and hence decohere).
Well, if one takes locality for granted (that's the point of course!), then one can use spacelike separation as a way to avoid environmental decoherence of the overall system (there will be local environmental decoherence, but it will not be the same environment degrees of freedom, and hence preserve the possibility of interference).
If one puts a macroscopic system in this way in a superposition (by having it observe one branch of an entangled system) in the basis of another macroscopic system (by having it observe the other branch of the entangled system in a rotated basis), and one brings then these two macroscopic systems together, one can try to find out if there is interference in their interaction, which would show up as correlations in their measurements. That's exactly what happens in an EPR experiment !
But, caveat: the only "superpositions" have been tiny electronic signals yet, over relatively small distances (50 km or so -> 10s of microseconds). That doesn't have much "gravitational" weight. We haven't really put human beings, or planets or so in superposition that way, for say, an hour or so. And there is still the issue of detection efficiency and the fair sampling hypothesis. So the "proof" isn't there yet that planets and so on can be in superpositions.
I'm focussing on the mass of the system in superposition, because I have to say that I'm sensitive to Penrose's argument. As long as we don't have a verified theory of quantum gravity, the possibility exists that superpositions of gravitationally important systems does not work, which would be the end of MWI. I fail to see how quantum computing proves this wrong.