Re: errata- Max entropy

[Leigh Palmer <palmer@sfu.ca>]

> Last night I was reflecting on what I wrote and realized an egregious mix-up
> occured.
>
> the section that read
>
> > The statistical underpinnings of the 2nd law of thermodynamics do not imply
> that the >entropy of an isolated system increases or stays the same for any
> given process.  It is >possible for an isolated system, through spontaneous
> fluctuations, to have a lowering >of its entropy.
>
> and below has a serious confusion.  Everywhere where I refer to an isolated
> system, I should have said a system held at constant temperature by means of
> being in thermal contact with a reservoir at that temperature.
>
> i.e. I was thinking in terms of a canonical ensemble, but wrote a
> description of a system best interpreted in terms of a micro-canonical
> ensemble.

What could possibly be the meaning of the entropy of *a* system which
is a member of a canonical ensemble? Are you suggesting that the
canonical ensemble is not itself a representation of *a* system?

I didn't comment on your earlier posting because I believe that there
is little meaningful content in the concept of an "instantaneous value
of the entropy of a system". Since so few seem to understand the
meaning of entropy in a sytem in thermodynamic equilibrium there is
insufficient context to make the more arcane concept meaningful.

If the microstate of any system is known to be any quantum mechanical
eigenstate whatever then the entropy of that system is simultaneously
zero and meaningless!

Chew on that one a bit.

Leigh