In my Self-Organization in Complex Systems talk in Siena for Georgi's CCS25 meeting September 4 and in the September 12 Evo Devo Scholar's Talk, I described Laplace's intelligent demon and Seth Lloyd's universe as a computer. They both view the information in the universe as a conserved constant.

We also discussed Lord Kelvin and Herman Helmholtz's prediction of a universe heat death.

Then I described my work following Arthur Stanley Eddington and David Layzer.

I showed how the expansion of the universe avoids the second law of thermodynamics and creates information structures from elementary particles to stars like our Sun and planet Earth with its biosphere.

I adapted the diagrams in Layzer’s 1989 book Cosmogenesis to create this picture of his growth of information in the universe.

Between the actual entropy and the maximum possible entropy you can see there's plenty of room for galaxies, stars, and planets to form.

The expansion of space creates new possible locations in phase-space, producing pockets of negative entropy. When an actual information structure forms locally, it will not be stable unless it radiates away positive entropy to satisfy the second law globally.

I call these two steps, first possibilities, then one actuality, the cosmic creation process.

These two steps or two stages are first indeterministic (random) possibilities, second an adequately determined (not pre-determined!) choice or selection.

This is exactly how Claude Shannon's theory of the communication of information works, showing the intimate connection between negative entropy and information!

Information philosophy proposes four such processes, all driven by random possibilities.

1. For Shannon, there must be multiple possible messages. If there is only one possible message, no information is communicated.
2. For the universe, without newly created phase-space possibilities, the universe would be closed and suffer a "heat death"!
3. Ernst Mayr called Darwinian evolution a two-step process. Without chance variations, there would be no new species.
4. For the two-stage model of human free will, without the mind producing random new thoughts, there would be no free actions.

The universe began with primeval quarks, gluons, electrons, and photons. In the first few minutes after the origin, the cosmic creation process produced the earliest information structures, protons and neutrons. 380,000 years later, the ionized plasma cooled to the surface temperature of the Sun and allowed those protons and electrons to form atoms, making the universe transparent. That allowed us today to see back in time to the cosmic microwave background.

Galaxies, stars, and planets began to form about 400 million years after the origin.

The Sun, a population I star, formed only about 4.5 billion years ago, along with its planets, and life emerged rather quickly about a half-billion years later.

We discussed how Erwin Schrödinger famously argued that life “feeds on negative entropy.”

Schrödinger’s source for negative entropy was our Sun. With the bright Sun as a heat source and the dark night sky as a heat sink, the Earth is a thermodynamic engine.

But Schrödinger didn't know how the Sun (and all the stars) came to be such a source of negative entropy. Following Eddington and Layzer, I have explained that with the cosmic creation process.

Twenty years later, Davies and two colleagues edited the 1994 volume Physical Origins of Time Asymmetry.

Davies contributed the article, "Stirring up Trouble." Without mentioning David Layzer's 1975 "Arrow of Time" article or Layzer's 1989 book on the Growth of Order, Davies coined the term "entropy gap" to describe Layzer's insight that the maximum possible entropy goes up faster than the actual entropy.

Does this transition from equilibrium to disequilibrium not constitute a violation of the second law of thermodynamics? No. What has happened is depicted in Fig. 3. At some time around one second, the material content of the universe was in a state of equilibrium, having the maximum possible entropy for the constraints at that time. As the universe expanded, however, the maximum possible entropy rose. The actual entropy also rose, but less fast. In particular, the relaxation time for nuclear processes to allow the cosmological material to keep pace with the changing constraints (due to the expansion) was much longer than the expansion time, so the material began to lag further and further behind equilibrium conditions ( equilibrium meaning in the nuclear case that this material is in the form of the, most stable element - iron). Hence an 'entropy gap' opened up. The continuing expansion of the universe serves to try and widen that gap slightly (though now through other processes than nucleosynthesis), while physical processes such as starlight production serves to try and narrow it.

It is important to realise that the crucial effect of the expansion was in the early universe - hence the sudden widening of the gap early on. Today it seems likely (though I haven't checked) that the gap is narrowing: the universe produces copious quantities of entropy at a rate which I imagine is faster than the (now rather feeble) expansion raises the maximum possible entropy. The actual entropy will presumably asymptote toward the maximum possible entropy in the very far future.

Physical Origins of Time Asymmetry, p.33

It's surprising that Davies suggests (though he hasn't checked!) that the entropy gap will close, leading to the 19th-century Kelvin-Helmholtz "heat death of the universe." David Layzer had no such pessimism.

Davies closing of the "entropy gap" suggests that the expansion stopping will make the universe a closed thermodynamic system!.