h2g2 The Hitchhiker's Guide to the Galaxy: Earth Edition

Probably you should never put a mathematician and a physicist in one room and let them talk about entropy.

Anyway, thermodynamics is a physical theory and it deals with large numbers of particles. Its definition of entropy makes sense only with this premise. Therefore the example with the coin is weak. By defining your instrument to measure 'not heads' you re-define the example in an unphysical way. Which instrument would say 'any temperature but not 3K'...

You can always translate thermodynamic facts into information theory. But I still think that both disciplines see it from different perspectives. Thermodynamics eg sees entropy only as so-called 'reduced amount of heat'. The whole theory doesn't contain the words 'disorder' or 'chaos'. However, this is the more interesting part, I admit.

The zeroth law is the definition of temperature; at least in thermodynamics. In the context of statistical physics, it's indeed a mathematical consequence, but this is true for most physical laws...

As far as the counterpart of the Third Law is concerned: I remember something called a Bolzmann Machine, many years ago. Maybe it has _anything_ to do with it. However on this field you are the expert.

Both, statistical physics and thermodynamics don't work with quantum mechanics. This may sound arbitrary but I must insist on that. With information theory you may do what you want tough.

Despite that, where can I read your quantum mechanics example? (We have a good library here.) It sounds interesting.

To get back to the roots, the Second Law is part of an approximative theory and thus can be violated, even without all these exotic thoughts so far. The smaller the scale, the more likely to be invalid.

Heehee. We've certainly cleared the forum of onlookers...

So effectively, thermodynamics is information theory, but applied only to 'large' systems, and hence it can say slightly more - but in a less accurate way...? That makes sense, I guess.

Why shouldn't statistical physics work with quantum mechanics? Seems like, since you can't describe the behaviour of individual particles, a theory to explain them in bulk would be exactly what you want? I'm sure there's a good reason - but could you explain it, or will it go _straight_ over my head?

Before the heads explode, something I heard from chemists some years ago.

The Fourth Law of Thermodynamics, the 'principle of maximal meanness': All processes develop in a way that I am most possibly annoyed.

The proof of the equivalence to Murphy's Law may be left to the reader as an exercise.

I always had trouble remembering the three laws until a lecturer at Napier explained them this way:

1st law says you can't win in thermodynamics; the best you can do is break even.

2nd law says you can only break even at absolute zero.

3rd law says absolute zero is unobtainable.

I guess this is accurate enough that someone knowing nothing about physics could get at least a fleeting idea of what the Laws of Thermodynamics imply, as long as he had no interest in understanding what the laws actually are.

The Laws are mathematical. In equation form, they're far simpler than the paragraphs you assigned to them. Their implications, however, are far-reaching, and include most, if not all, of your article. But suggesting they mean so little does the subject itself a disservice.

Thermodynamics is by far the most complicated scientific topic I've ever studied, and I understand very well that anyone studying it for less than a year or two would have a plethora of mistaken assumptions. Further, I understand that "dumbing down" (no offense to fellow readers) these concepts for those with little or no science background requires introducing a lot of misleading and inaccurate statements in the interest of giving the Big Picture.

A simpler and much more accurate summary of the laws is thus:
1. You can never get more energy out of a system than you put in.
2. In fact, you can't even break even.
3. Well, you could break even at absolute zero, if you could ever attain it. Which you can't.

I guess my peeve here is that I've read about a half-dozen science articles on H2G2 in the past few days, and every single one is laden with inaccuracies. I worry about people who use this as a scientific resource and fill their heads with wrong ideas.

Oh, and thermal equilibrium is dependent on temperature, not "heat energy," whatever you think that is. Your statement is so terribly wrong I want to scream.

Alchemy

Life does not violate the 2nd law, as it exists as part of a larger system, ie: the universe. The 2nd law specifically holds for isolated systems- those that have no energy or mass losses or gains from outside the system. The Earth is clearly not an isolated system, as we gain both mass and energy in rather large amounts on a daily basis. The 2nd Law (and all the other ones) are absolute, if the system is isolated. That said, they still hold well for other system types