Re: Meta-classes? (layered architecture)

On October 16, pat hayes writes:
> >Firstly, my main point (with which I think Pat agrees) is that we
> >should define a simple "base" language, and then extend it as
> >necessary with -META -SECOND_ORDER or whatever.
> 
> Yep, I think everyone agrees with this, more or less.
> 
> >My particular worry was the potential confusion that is derived from
> >building on top of RDFS, which is designed so as to be able to
> >describe itself. For example, according to the specification,
> >rdfs:Class is a type of rdfs:Class as well as being a subClassOf
> >rdfs:Resource, which itself is a type of rdfs:Class. An rdf:Property
> >is also a type of rdfs:Class.
> 
> Ah, I may have been misreading this.  Maybe it is my turn to 
> demonstrate lamentable ignorance in public, but could you briefly 
> explain what the distinction is between being a type of and being a 
> subclass of? I assumed that the fact that rdfs:Class is a type of 
> rdfs:Class means that classes are a kind of class, which seemed 
> harmless since it is a tautology. (Perhaps I should understand the 
> first occurrence of 'rdfs:Class' here to be a mention rather than a 
> use?)

My understanding is that for "type" we can read "instance of". e.g.,
rdfs:subClassOf is an instance of rdf:Property, not a kind of
rdf:Property.

Ian