Both the “entropy is in the mind” and “entropy is real” explanations seem plausible to me (well, I am not a physicist, so anything may seem plausible), so now that I think about it… maybe the problem is that even if we would be able to know a lot of stuff, we might still be limited in ways we can use this knowledge. And the knowledge you can’t realistically use, it’s as if you wouldn’t even have it.
So, in theory, there could be a microscopical demon able to travel between molecules of boiling water without hitting any of them—so from the demon’s point of view, there is nothing hot about that water—the problem is that we cannot do this with real stuff; not even with nanomachines probably. Calculating the path for the nanomachine would be computationally too expensive, and it is probably too big to fit between the molecules. So the fact is that a few molecules are going to hit that nanomachine, or any greater object, anyway.
Or perhaps we could avoid the whole paradox by saying: “Actually no, you cannot have the knowledge about all molecules of the boiling water. How specifically would you get it, and how specifically would you keep it up to date?”
This is pretty much it, and it’s a really subtle detail that causes a lot of confusion. This is why the real problem with Maxwell’s demon isn’t how you obtain the information, it’s how you store the information, as Landauer showed. To extract useful work you have to erase bits (‘forget’ knowledge) at some point. And this raises the entropy.
Both the “entropy is in the mind” and “entropy is real” explanations seem plausible to me (well, I am not a physicist, so anything may seem plausible), so now that I think about it… maybe the problem is that even if we would be able to know a lot of stuff, we might still be limited in ways we can use this knowledge. And the knowledge you can’t realistically use, it’s as if you wouldn’t even have it.
So, in theory, there could be a microscopical demon able to travel between molecules of boiling water without hitting any of them—so from the demon’s point of view, there is nothing hot about that water—the problem is that we cannot do this with real stuff; not even with nanomachines probably. Calculating the path for the nanomachine would be computationally too expensive, and it is probably too big to fit between the molecules. So the fact is that a few molecules are going to hit that nanomachine, or any greater object, anyway.
Or perhaps we could avoid the whole paradox by saying: “Actually no, you cannot have the knowledge about all molecules of the boiling water. How specifically would you get it, and how specifically would you keep it up to date?”
This is pretty much it, and it’s a really subtle detail that causes a lot of confusion. This is why the real problem with Maxwell’s demon isn’t how you obtain the information, it’s how you store the information, as Landauer showed. To extract useful work you have to erase bits (‘forget’ knowledge) at some point. And this raises the entropy.