[topicmapmail] Subject Identifiers metadata

[Thomas B. Passin]

Dan Corwin wrote:
> Thomas B. Passin wrote:
> 
>> I don't agree about "PSIs seem unlikely" in this connection, though.
>  
> I was reacting only to your earlier suggestion that PSIs might work as
> a way to let one refer to topics in a TM as subjects in their own right.
> 
> My point was that if those topics are being created *dynamically* in
> bulk by a running program, they are unlikely to have any PSIs.
> 

Ah, I see.  No, I meant that there would be a PSI for the _type_ of the 
topic that represented a subject indicator or another topic.  That would 
be one PSI for a type that represents a subject indicator, one more for 
a type that represents a topic, and so on.  Just a few of them.

>  
> A more urgent concern to me is that I cannot reify a topic TODAY under
> XTM 1.0 specs and current engines, or at least I do not know I can.
> 
> I like your suggestion on using <subjectIndicatorRef> in XTM 1.0, but
> would it actually work?   Hence my Q3.
> 

Well, it depends on what you mean by "work".  Mechanically, you can do 
it.  Whether a general purpose topic map system will do anything special 
with these constructs is another question.  But you will always be in 
that position when you invent a new type.  The engine will deliver to 
you the topics or whatever that you ask for, but you may have to write 
code to do the appropriate thing with them.

It's like OWL and RDF.  If an RDF processor ingests some OWL, it can 
construct the basic triples declared in the RDF document, but it does 
not know to do anything OWL-like with them.


>> Make sure not to mix up the idea of reifying a subject indicator with 
>> the idea of referring to a specific topic.
> 
> 
> Exactly right.  Kal's original post on this thread suggested there were
> multiple ways to the former, and he only wanted to explore the best.
> 
> I muddied up the waters (sorry, Kal) by expanding the scope of the issue
> to reifying topics as well, which may or may not closely relate.
> 
> Chreers,
> Dan
> 
> PS - Thanks for your other comments, Tom.  I'll respond separately to
> them to avoid additional "scope creep" in this thread.
> 

Cheers,

Tom P