Tommi

Why does everyone seem to insist that Maxwell's demon is an ideal and arbitrarily small machine? Wouldn't a passive element be much more natural to consider? Further, why is it nowadays considered, or likened to, a physical existence, when it is literally a demon? In control of a wall with no physical qualities beyond reflecting particles? Wasn't the "agent" described as a demon because it did not physically exist in the speculation. Couldn't, and shouldn't, the opening and closing of a spooky wall be transformed into a series of diffraction elements, lenses if you will, that redirect a particle to a nearly 90° angle, if it does not approach at such, followed by a nonabsolute barrier, a potential energy slope if you will, that can only be surmounted by a particle of sufficient speed. A failed particle would reverse motion only perpendicular to the slope, and is redirected by the "lenses" to a new trajectory, as if it was somehow delayed from ricocheting. Without the delay, the motion of the particle against the slope would be wider, and the barrier would as such lose efficiency to the lateral velocities of the particle. Like this, the thought experiment is transformed from superstition into science fiction, and the only spooky things that remain are the two passive elements properties as perfect insulators. I'm fairly sure diffraction does not qualify as spooky. It was a thought experiment of how to violate entropy, a simplification of the process, it should not be depicted in a manner that straw-mans any notion defying the law.

Would you believe it's the first time I hear internal space being specified? Because I can't remember hearing about that before. Still, wouldn't measuring the particle along a specific axis effectively polarize it in ordinary space as well? The obvious part I meant in a more general sense, I'm afraid. I'm thinking about the polarization demonstrations with rope, where a circular wave is converted by a slit and negated by the next which is at 90º to the first. It seems perfectly intuitive to me that if a non-orthogonal slit is placed between the two, it's going to change the wave that passes through and let more through the next. What I'm asking is whether or not this interaction with the 45º is something relevant, or that should be compensated for with it's own orthogonal pair.

So, it's terribly obvious to me that a, say, 45 degree angle between the two orthogonal ones is going to let you get more light through, the same applying to all polarized waves. Which is why the inequality is violated often, as a non-90 degree angle changes the polarization. What is less obvious to me is what happens if you have a fourth direction of measurement, say again, at 45 degrees to the two orthogonal ones, but which is also orthogonal to the third, and no measurement is taken on a shared direction? What I mean is a second pair of orthogonal directions at an arbitrary angle to the "original" pair, say z & x & q & w, and just as it is pointless to measure z twice, it should be pointless to measure q twice, so you measure w instead. What does this do to the inequality? Or is this a pointless question and why?