In the Bayesian and inductive-logical view of probability theory, you must define a probability distribution with respect to an information set. That set formalizes your observations of (and assumptions about) the state of the universe, and the entropy of the resulting distribution quantifies the quality of your knowledge about the universe.
From that perspective, the second law of thermodynamics simply says that the quality of your knowledge about the universe with respect to an information set assembled at time t degrades as you move from t out into the future. Put like that, it seems rather obvious, I think.
Quantum theory further tells us that it's not possible to assemble an information set that is complete, in either the sense that (a) we can get a perfect picture of the universe as it is right now, or that (b) the quality of our information won't degrade over time. (The two senses are equivalent, and you can probably see why.)
So Laplace's demon can't exist. Which isn't terribly surprising to the modern mind, either, but I can imagine how scientists of a previous era might have taken the news badly.
From that perspective, the second law of thermodynamics simply says that the quality of your knowledge about the universe with respect to an information set assembled at time t degrades as you move from t out into the future. Put like that, it seems rather obvious, I think.
Quantum theory further tells us that it's not possible to assemble an information set that is complete, in either the sense that (a) we can get a perfect picture of the universe as it is right now, or that (b) the quality of our information won't degrade over time. (The two senses are equivalent, and you can probably see why.)
So Laplace's demon can't exist. Which isn't terribly surprising to the modern mind, either, but I can imagine how scientists of a previous era might have taken the news badly.