[topicmapmail] Subject Identifiers metadata

[Dan Corwin]

* Kal Ahmed:

>>>I was talking about the thing that we conceptually use to represent a
>>>subject - irrespective of its representation in some computer system.

* Dan Corwin:

>>If there truly were no such subject, Kal, then how can you discuss it
>>so easily with us in both of your paragraphs above?

* Kal Ahmed:

> What could you possibly want to say about it?
...
> I can't see what you can possibly say about the "thing that we hang
> assertions off of" - its essentially a non-entity that comes into being
> only when we make an assertion about a subject.
> 
> There may be some philosophical point to wanting to do this, but its not
> within the reach of my developer-brain ;-)

Already asked and answered once.  If I can clarify any of the half dozen
specific, non-philosophical examples I gave, please let me know which:

   http://www.infoloom.com/pipermail/topicmapmail/2004q2/005986.html

Your developer-brain may better grasp my intent by reflecting :-) on
Java's similar powers of introspection, which lets one *use* Java
objects to *represent* (within the Java engine and paradigm)
other classes and objects modeled AS DATA STRUCTURES.

Analogously, I'd like to *use* XTM's topics to *represent* (within
the Topic Map engine and paradigm) other topics and their types,
also modeled AS DATA STRUCTURES.

Admittedly, there are issues lurking here, for data structures and
their parts are not trivial things to formally identify and model.

But Tom and I agree on a simple and obvious convention that helps
support such work, which appears to require no changes at all to TM
syntax or engines, and hence exists purely at application levels:

   http://www.infoloom.com/pipermail/topicmapmail/2004q2/006003.html

Regardless of *WHY* we might want to do it, do you see any practical
barriers at the levels of TM4J or XTM 1.0 about *HOW* this works that
might limit its adoption today by "introspective" applications?

Please note that two answers may be needed, as this proposed method
seems to allow adding arbitrary metadata to BOTH subjectIndentifiers
(your issue) AND to any other topics (mine).

The only obvious differences I see, philosophical or otherwise, lie in
which declared sub-type of DATA STRUCTURE one wants to reify, which
(perhaps) affects how its specific examples get identified and found.

If that difference has consequences in the TM4J world, please be kind
enough to explain.  Thanks.

Cheers,
Dan