Quote by vanesch
Well, Penrose's not so enthousiastic about inflation solving this problem and I have to say that I find his argument convincing. The reason is that thermal equilibrium (high entropy) is NOT equivalent to homogeneity when gravity is taken into account. Homogeneity is a state of LOW entropy (far from thermal equilibrium) when gravity is taken into account, and as such, using a timereversible mechanism such as inflation to explain a LOWentropy situation does only report you to a _still more stringent_ condition before it. You cannot have "matter thermalize to give you a uniform distribution" on a small scale (unless you *switch off gravity*). If it were to "thermalize" (with gravity) it would generate lots of singularities, and that wouldn't give rise to a smooth uniform homogeneous structure after inflation. The only way inflation can give rise to a (low entropy) state of homogeneity is that there was a potentially even lower entropy state before it acted.
Now, I'm not an expert on this stuff at all, but I found this argument extremely convincing  although I can understand that it must be somehow controversial.

This seams to be something profound but I cannot follow it... Black holes have a great entropy, but it seams to me that they are not the states of greatest (total) entropy because they do evaporate. When a black hole evaporates, it converts the Schwarzschild spacetime into another spacetime. What is the resulting spacetime? My first guess would be that it is a (expanding) space with a homogeneous distribution of radiation. In such a case, how can we say that homogeneity is a state of low entropy when gravitation is taken into account?