[tm-pubsubj] Subject Indicator vs Subject Descriptor

Following November 20 meeting, consensus of present people 
seems to be established on the following points (correct me if I am 
wrong). 

A Published Subject should provide two distinct parts:

1.  Subject Indicator (URI or URL)
as to be used through <subjectIndicatorRef xlink:href>
It is thought that this should be both 
-- a syntactic identifier, to be used by software as a mere character 
string, without any further processing and interpretation. 
-- a pointer to a resolvable resource (the Subject Descriptor defined 
below). 
No, or the least possible, "semantics" are to be put in the string 
itself, apart from a standard structure, allowing software or humans 
to identify it as a Subject Indicator.

2.  Subject Descriptor
It is the content of the resource that the Subject Indicator resolves 
to. It is where the subject is properly defined and/or described, in 
other terms it is where the semantics are grounded, and the Topic 
Map recursive process ends. 
For the recursivity to end effectively, the Subject Descriptor have to 
be defined not in Topic Map terminology, but at a meta-level.
Best candidate for standard Subject Descriptor structure could be 
built upon RDF-Dublin Core elements.
Such a standard Subject Descriptor could be used by parsers to 
extract metadata on the subject, like publisher, date of creation, 
validation etc.

I propose that this distinction be explicitly stated on TC 
requirements, as well as the requirement to deliver a syntax for 
both Subject Indicator and Subject Descriptor.

Comments *most* welcome.

Bernard
Bernard Vatant - Consultant

Mondeca - "Making Sense of Content"
www.mondeca.com