That's interesting, but also a bit maddening...

...since he never defines "entropy" as he understands it. There are plenty of spots where he says that entropy isn't something, but not one where he says just what it is. He comes close when he cites Clausius:

"Finally, in 1865, he coined the term 'entropy,' writing, 'I propose to call the magnitude S the entropy of the body, from the Greek word τροπη, transformation.'"

It seems like the biggest problem is that the term "entropy," which was devised to express something about the state of a thermodynamic system, has come into use outside of thermodynamics. See link for another kind of discussion.

Some day I'll try to convince our younger daughter to write up her theory that we use the term "random" very loosely, and that even random number generators don't operate randomly. (The literature recognizes pseudo-random generation and "true" random generation, but her argument is that even the latter is random only because we can't predict the inputs, which would be predictable if we knew more about them.)

Click here for link at https://www.panspermia.org/seconlaw.htm