Nostr Web Client

I think the 2nd law of thermodynamics and the idea of entropy is one of the most confused and misunderstood ideas in physics, and the fact that we have accepted this handwaving argument for so long is shameful.

And I'm a credentialed computer scientist. Entropy has an informational meaning. I'm well aware of Claude Shannon's work. But let's go back to physics for a second.

Entropy is a measure of disorder. But what is disorder? What is order? There is no such thing. It is a personal preference. I like my pot on top of my stove, but my girlfriend likes it in the cupboard. Who is right? Which state is more disordered? There is no objective answer.

If there is no objective definition of which states are ordered and which are not ordered, then there is no true entropy.

All we actually have is the law of large numbers. If you choose to concieve of a small set of states among the nearly infinite possible states, then the odds that something will move from one of those infinite states into your small set is astronomically low, compared to things moving in the other direction. But as the lord Buddha says, empty your mind. Dont concieve of a small set of states. Where did the entropy go? You created it! The entropy was all in your head.

Reply to this note

Please Login to reply.

Discussion

Mike Dilger ☑️ 1y ago

Someone might object and say something about the likelihood of various states. So then I have to explain what probability actually is.

The universe only happens one way. If I toss a penny, it lands either heads or tails. Let's say I toss a penny and catch it and I peek at it but I don't show you. I ask you what are the odds it is heads up? You say 50% and I say "Nope. The odds are 0%" because I cheated, I saw that it was tails up. So what were the odds _really_? Objectively? That is undefined. Probability isn't objective. It is not just about the universe, but about how much you know about that universe, and much more so about you than about what you think it is about. To keep this clear, I like to think of probability as the measure of your ignorance.

Mike Dilger ☑️ 1y ago

And if you thought that was at all interesting, listen to Sabine. I think I agree with her. https://www.youtube.com/watch?v=89Mq6gmPo0s

to be decided 1y ago

Probability might not be defined objectively in the specific case of your coin toss experiment. But the coin toss is a very high level experiment.

In respect to our shared material world (which is governed by low level physics), all of us have the exact same amount of ignorance (noone can observe the microstate of matter in a sufficient manner for it to make any difference in comparison to other observers) and therefore must assume the exact same probabilities of micro states, which does make these probabilities objective indeed in practice, doesn't it?

Mike Dilger ☑️ 1y ago

Functionally speaking, yes. Fundamentally I don't think so. Even though we know how to pull work out of a system via macrostate order and can't from mixed systems due to our microstate ignorance, I don't think that means that it is impossible for anything to have enough microstate knowledge to derive useful work from it.

I don't think it's something we are going to solve soon. It is almost as crazy of an idea as hoping for perpetual motion, time machines, or wormholes in space. I don't hope for it ... I just think we haven't proven that it is out of the realm of possibility.

to be decided 1y ago

> .. I don't think that means that it is impossible for anything to have enough microstate knowledge to derive useful work from it.

I obviously cannot fathom how this could ever be possible.

On the other hand, sometimes my mind takes this non natural sciences angle on things for a second and I just think: considering how crazy/curious/strange/weird/beautiful the mere fact that "things exist at all" is, it sure wouldn't be that much more crazy if things like wormholes or telepathy existed as well.

BUB 1y ago

You've created a straw man regarding the 2nd law of thermodynamics and entropy.

a very simplictic idea about "order" using semantics.

BUB 1y ago

If you want to reverse a process the entropy increases.

No thermodynamics course uses language like order and disorder. In anything more than a passing example.

Mike Dilger ☑️ 1y ago

I'm not denying any of that. I'm pointing out that a part of the definition is subjective. See my further on comment linking to Sabine Hoffsteader's video. She talks for most of the video about all of the things I'm sure you understand about entropy, but fundamentally seems to agree with me at the end.

to be decided 1y ago

> Entropy is a measure of disorder. But what is disorder? What is order? There is no such thing. It is a personal preference.

I guess you're familiar with the following example: One container with two different gaseous substances in it, both 99,99 % pure, separated by a barrier. You open/remove the barrier and the gases will mix by diffusion.

By your personal definition of the word "disorder", these two situations (before and after opening the barrier) cannot be distinguished by the measure of disorder?

If you take two containers, one with blue lego parts, one with red lego parts and you pour both into a third container and shake it for a minute. It would align with your understanding of "disorder" to say: "The mixture of both blue and red has the exact same measure of disorder as when both colors were separate."?

Mike Dilger ☑️ 1y ago

I'm saying that disorder is defined by which states you feel are ordered. If you feel that having all one gas on one side and all another gas on the other is "ordered" then yes removing the barrier increases disorder. My point is that you had to choose a small set of possibilities and call it "ordered" which makes the notion of entropy partially subjective.

Viktor Vsk 1y ago

“A small set of possibilities” in that case is when you mix 2 lego boxes each with its own orders, mix it for a minute and get 2 layers one completely red and other one completely blue - thats what a small set of possibilities mean in this case

Otherwise probability is always 50/50 either yes or not