Time Asymmetry in a Nutshell

The follows is my first attempt at summarizing the search for the origins of the second law.

Entropy must start out small at beginning to observe non-equilibrium now. Given current observations, the universe did just this. Low entropy corresponds to a smooth universe in a system dominated by gravitation which we observe in the CMB. So the question we are asking is, how did it get this way? Is there a dynamical reason or just some initial condition?

Arguments usually center around coming up with some dynamics that allow generic initial conditions to evolve to what we see today.
One of the first proposed that inflation itself was responsible, simply flattening out all the inhomogenaities is its reason for existence anyway. But when gravity is repulsive, flattening inhomogenaities is still increasing entropy. (Albrecht and Guth)
One of the last arguments on the table is that the equilibrium state is actually unstable to inflation. Take a stable expanding universe, add an inflaton field and a zero point field and you can get thermally driven spontaneous inflation. This inflation can be "eternal" with "pockets" of reheating which look like big bang universes. (Carroll et al)

Need to read: Eternal inflation , inflaton fields, de Sitter space details.
Next time, more readings.

Old Idea - New Idea

When I first read about the Hubble expansion many years ago I wondered if it would be possible to build an experiment that directly measured the effect. I pictured some kind of orbiting laser interferometer that measured the relative redshift between two widely spaced points.
Now, for some reason, I remember this today and I think, "well here is LISA, which is exactly that". Where does the Hubble expansion lie on one of those signal vs frequency graphs that gravity telescope people are so fond of showing.
If H_o is approximately 75 km/s/kpc than the path length difference between two LISA endpoints is 4x10^-20 meters. Now I just have to understand the language of the science, do they talk about strain (h_+ etc) or path length difference, something even more obscure?

More: (20/4/05)
The LISA system will supposedly be able to measure a phase difference between two arms to one part in 10^20. The hubble redshift over the length of one of LISA's 10^9 meters will be 10^-15. Does this mean we will see a strong DC signal from the expansion?

The hubble expansion would represent a DC (not counting cosmological acceleration) signal in the redshift and a linear signal in the distance. I dont really know what the output of a time delay interferometer (TDI) is. Does it integrate for distance with some analog/optical device or do we get a stream of phase differences over time? They talk about strain spectral density, which has units of 1/(Hz)^1/2, thus total strain might have units of Hz^1/2....