Comment by ncallaway | Hacker News Reader

Comment by ncallaway | Hacker News Reader

Parent Full thread

The general idea of Maxwell's demon as a problem is that it conflicts with the second law of thermodynamics, which essentially states that the total entropy of a closed system will increase over time.
Which, essentially, implies that if we have a (perfectly sealed closed system) box of gaseous particles well mixed at temperature T, then we will never observe that box eventually separates into two different temperatures of gas with a T(cold) on the left side of the box and a T(hot) on the right side of the box. However, if we already had a box that was separated into two different temperatures of gas, we would expect those gasses to eventually become well-mixed over time.
The paradox of Maxwell's Demon was that we had this apparently closed system (gas particles + Demon), that wasn't doing any Work, but apparently had the ability to violate the 2nd law of thermodynamics. The thinking was that this shouldn't be possible, even in principle.

With the caveat that the second-law of thermodynamics is a statistical law, so it's really saying "the odds of entropy decreasing by any significant amount is extremely vanishingly improbable"

What bothers me is, we derive entropy --a blind, involuntary physical law--, from observation of inanimate systems.

So, for example, a very big box full of hot hydrogen will eventually cool off. But, there's a tiny chance it may actually coalesce into life bearing stars. Such life may intentionally work towards increasing order, like Maxwell's demon.

So we try to extend entropy to life too and strenuously prove that it'll increase even when aware biological organisms work against it, while still excluding the same organisms from the observations entropy is derived from.

Are there no natural non-biological processes which decrease entropy locally just as life does? Surely something like a weather system does that, doesn't it?

Crystals might be another example.

Talking about physical entropy (Boltzmann) or information entropy (Shannon), in either case we must consider the system or channel.

Entropy only has meaning in respect of a system. There is no "universal* application of the second law.

Entropy within a system may very well decrease so long as more entropy is exported from that system.

Living organisms import low entropy such as food. And they export higher entropy.

The balance in a notional system that encompasses both the organism and its energy source, wrt beyond that system, will be a net inrease in its entropy.

It's easier to understand with a refrigerator.

A refrigerator decrease the entropy inside (ice cube have low entropy) but it increase the entropy outside, so the total amount of entropy in the universe increase.

There are even some refrigerators that instead of electricity have a burner for alcohol, that for me is very weird.

Now imagine a big box filled with Hydrogen an a refrigerator (with batteries or something). The entropy inside the refrigerator will decrease, but the total entropy will increase.

Now imagine a that there is a star in the big box of Hydrogen, and the refrigerator has a solar panel. The entropy inside the refrigerator will decrease, but the total entropy will increase.

Life is just that, a weird solar powered refrigerator.

Reply on news.ycombinator.com