Malconstant wrote:But the demon need only create my experience via matrixesque simulation, and to hell with the rest of the universe. So if the demon itself has a much simpler universe than the fake one being modeled, there you go. Does that really feel more complex to you than our infinite universe possibly multiverse with qm, relativity, brane theory, 11-dimensional string theory, etc., etc. Maybe our demon has a much simpler universe but a very creative mind.
Also with "measure" and "Boltzman brains" I was invoking a probability argument. What's simpler, a universe which appears to have evolved for billions of years to make everything as we see it, including human civilization. Or, that my brain and perception is a random fluctuation in a suitably large bath of hot plasma, and in just one half of a split second the surrounding plasma will flood my fluctuation and annihilate me. For that one all you need is randomly fluctuating stuff lying around long enough, is that not simpler?
Ah, but for the demon to create the phenomena that we observe it has to imagine them, and therefore the universe in which the demon exists has to contain at least as much information as what ours appears to have, plus whatever extra information doesn't get through to us. You can't simulate a complex system within a simple one without increasing the complexity of the simple system to at least as high as the system you're trying to simulate. So even if the demon's universe operates on very simple laws which result in complex phenomena, those simple laws need to be at least slightly more complex than the simple laws we could hypothesize to govern our own universe.
Boltzman Brains, on the other hand, offer a much simpler generative
system for the universe - unpredictable fluctuations. Crucially, however, to compress the data present in a series of unpredictable fluctuations is actually harder
, since you can't reduce them to simple laws (if you could then they wouldn't be unpredictable). So either we can posit that Boltzman Brains to arise as a result of predictable fluctuations, in which case the universe can be reduced to simple laws (we just won't be able to discover them easily) or we can suggest that the fluctuations are unpredictable, in which case it's an even more complex hypothesis again.
Of course, it's totally arguable that the behaviour of the universe isn't
reducible to simple laws, but doing that is begging the question, since it presumes that Occam's principle is wrong.