Brian
I like what you did.  My main suggestion is pushing an overview (the hierarchy)
to the very beginning.  I interspersed my notes enclosed in <<>>.
I am using my definition of rdfs:Class which nobody has blessed.

RDF Classes

Resources may be divided into hierarchical groups called classes.
<<With RDFS Classes, it's really a lattice, but: (a)that's too much technical
detail for an introduction; (b)many people don't even realize that there's
a difference between the two -- they use the word "hierarchy" when
referring to a "lattice".>>
The RDF Schema classes include:

	rdfs:Resource
	    rdf:Property
		rdf:type
		rdfs:subClassOf
	    rdfs:Datatype
	    rdfs:Class

Each member of a class is a resource, and each class is itself a resource.
The members of a class are sometimes called instances of the class.
<<I recommend against using "instance", but if you do, this statement is good.
On second thought, "instance" is so prevalent now, it's probably a good idea
to use this, even if you never use the word "instance" again.>>
In different contexts, two classes may have exactly the same members and 
yet be different classes.

Resources are often identified by URIref's [@@ have we defined the
rdf use of this term yet?] and may be described using 
RDF properties.

rdfs:Class is a "metaclass": it is the class of all class names.
<<I originally had "very special case" instead of "metaclass".
I think "metaclass" is an improvement.>>
<<this statement is false: The group of resources that are RDF Schema
classes is itself a class called rdfs:Class.>>
This may seem a little confusing to some, and especially to 
those who think of an RDF Schema class as a set of resources, for the most 
familiar set theory teaches us that sets cannot be members of themselves. 
However, as explained in [@@RDF-SEMANTICS], a little mathematical trickery 
enables RDF Schema classes to behave like sets whilst still enabling them 
to be members of themselves.
<<It may be good to leave this in -- for now -- it's a good statement
of Pat Hayes' position.  But Pat is wrong.  Pat is combining two different views
of what a class is, i.e., combining two different definitions which apply to
two different contexts.>>

A class A is a subclass of a class B if and only if all the members of A 
are also members of B.
All classes are subclasses of themselves.
<<this is true, but RDF Schema should use proper subsets instead of subsets.
Using subsets logically permits such absurdities as: the set of all men is
identical to the set of all animals.>>
The rdfs:subClassOf property may be used to state that one class is a subclass 
of another.
The rdf:type property may be used to state that a resource 
is a member of a class.

The RDF Concepts and Abstract Syntax [@@RDF-CONCEPTS] specification defines 
the RDF concept of a datatype. RDF Schema defines all datatypes to be 
subclasses of rdfs:Datatype. The members of each subclass are the members of
the value space of the datatype.


The core RDF specifications define the following classes.

rdfs:Resource

All things described by RDF are called resources, and are members of the 
class rdfs:Resource. This is the class of everything. All other classes are 
subclasses of this class. [@@ is this true; is it stated or implied in the MT?]
<<this is true>>

rdfs:Class

This is the class of all class names.  The name rdfs:Class is a 
member of rdfs:Class.

rdfs:Datatype

rdfs:Datatype is the class of datatypes.
<<not true: It is a subclass of rdfs:Class.>>
<<My concept of rdfs:Datatype is still fuzzy.  I'm not sure if it's a subclass
of rdf:Property.  I have frankly ignored Datatype writeups in favor of
figuring out the meaning of rdfs:Class.>>

....