[:] J Schnick wrote in part:
The other question can be exemplified by a rotating bullet. It seems reasonable to consider the bullet to be the system. And it seems reasonable to create a fixed imaginary boundary about a region in space that is at all times occupied by the bullet and call that region the system. But I am not familiar with the idea of calling the spin of the bullet a subsystem and what is not the spin of the bullet another subsystem. It seems that each subsystem should be a system itself. I can see considering the bullet to consist of two parts, a nose part and a tail part for instance, and each of those parts can be a subsystem, but I'm not familiar with the way you used the term subsystem as a subset of the degrees of freedom. Is that a common use of the expression?
[:]
I"ll take a stab at the last paragraph? "Subsystem" means partitioning a system in some meaningful sense. I can partition by sub-volumes if I want; (the nose of the bullet vs the tail of the bullet. I can partition by sub-masses. But there are many other possibilities, including partitioning by other categories; even more abstract than volume of mass. For example, I could partition by modes of energy with differing occupation numbers; in which the bulk spin mode for the bullet is in a different partition (sub-system) than the partition that includes vibratory KE modes for localized atoms in the bullet, which is in a different partition than PE vibratory modes of the bonds between the atoms, etc.