Basically, the important consequence is that different branches wouldn't evolve in isolation any more. They'd interact. To be fair, that's why it's called an "interpretation". Also it's not completely unfalsifiable: if quantum theory were ever found experimentally to be inadequate in some way, the MWI would fall with it.
Yeah, it seems to usually be presented that way, but the kicker is that it uses a bunch of assumptions of CFD in the process. And as MWI proponents later pointed out, you have no such problems with locality if you don't start by assuming CFD. So while Bell thought that he was proving that QM was nonlocal, it was later understood that he was only demonstrating that locality AND CFD don't play well with QM. Or that's my hand-wavey understanding anyway. Right, as I understand it the "Bell inequalities" themselves don't apply in MWI since they require CFD. But if you go and reformulate them under the assumptions of MWI, your results match up quite nicely to observation without any need for nonlocality. This is why it's said that Bell's Theorem requires us to ditch either locality or CFD - it's not the inequalities themselves, but their applicability, that gets the larger result. In hand-wavey terms, CFD is required for those integrals you cite, since they involve summing over events that aren't observed to occur. From the MWI perspective, Bell errs by assuming that the entire universe is in a single definite state once the first measurement has been made (should be easy to see why that doesn't mesh with locality). MWI does not: each measurement is a split of its own, and they are only brought into alignment much later, when (if) the two outcomes are transmitted to a common observer (at a speed below that of light), at which point there is another split that rectifies everything. Anyway, here's a more detailed treatment of how MWI and Bell relate (annoying ascii-math, but pretty interesting regardless): http://www.hedweb.com/manworld.htm#bell
Yeah, exactly. These two raise a related point, which is that any meta-theory ends up becoming an actual theory if you press it into areas the original theory didn't cover. When it comes to vanilla QM, MWI doesn't have anything falsifiable to say. But if one wants to keep MWI in the case of quantum gravity, one has to either predict that quantum gravity is linear (this is falsifiable) or accept that there are observable interactions between decohered worlds (this may or may not be falsifiable, but in either case represents a radical shift in the picture of the universe MWI is supposed to exhibit - i.e., that it's only one world with many temporary local branches).
'the "observer"' is simply widening the 'conversation' about the experiment. How 'the "observer"' acts, in the experiment, is just another "probability", which you have added (introduced) to the equations. The Double slit experiments, demonstrates the wave-particle duality of the electron. "Starting with an electron 'particle', which then passes through two slits, acts like light waves, forming an interference pattern." To go deeper into the experiment and theory, now ask yourself, "What happens if you remove the electron 'particle' completely" You don't change the experiment. From electrons to black holes and now even Prof. Susskind, all quantum tunneling through to cause an interference pattern. Who would have predicted that? Please Register or Log in to view the hidden image! Please Register or Log in to view the hidden image! Please Register or Log in to view the hidden image! My suggestion is that everyone watches the Richard Feynman and Leonard Susskind lectures, with an open mind, and learn from the masters. Light in, Light out.
