On 6/1/17 6/1/17 3:36 PM, Phillip Helbig (undress to reply) wrote:
In article ,
Gerry Quinn writes:
[...]
Someone has made an error somewhere.

Yea. I believe it is in your entire approach: I don't think that
entropy is an absolute quantity as you implicitly assume.

It does make sense to compare entropy between states of a given
universe, but not between states of different universes. After all,
entropy is an extensive quantity, and in considering different
universes you cannot possibly ensure they have the same values of
all other extensive quantities.

In "changing the value of G" you really have a different universe
for each value. You cannot change the laws of physics in a universe,
you must consider an ensemble of universes.

For instance, consider two universes, one with G=0 and one with
G0. Give them the exact same initial conditions, in that the initial
geometries are the same, as are all field distributions and
derivatives. Since the Lagrangians are different, the evolutions
of the fields and geometries will be different, the total energies
will be different, etc. -- such extensive [#] differences surely
invalidate any comparison of their entropies.

This presumes it is possible to give them the same initial
conditions. It is not obvious that this is so.... Certainly GR
has strong constraints that the initial conditions must meet;
different values of G could yield different constraints.

[#] This is a pun: "extensive" as in type of property, and
"extensive" as in vast or widespread.

Tom Roberts