[topicmapmail] Re: topicmapmail digest, Vol 1 #1104 - 8 msgs

[Michael Della Bitta]

On Sunday, September 28, 2003, at 09:07 PM, Jan wrote:

> No. The meaning of a URI is not dependent of the context in which it
> is used. The authority controling that URI **defines** what it denotes!
> Crazy, yes. But that is how it (currently) is.

Frankly, the context of *anything* is dependent on the context in which 
it is used. Such is a symptom of the postmodern world.

This is a classic example of the sense and reference distinction from 
the philosophy of language: meaning can easily be subdivided by a. the 
thing being referenced (e.g., the output at the end of a URL), and b. 
the sense in which it is being used (e.g., the tag the URL is an 
attribute value of).  Just as I may make mention of 'the morning star,' 
and the sense that that term takes it's meaning from diverges from 'the 
evening star,' even though astronomically, they reference the same 
thing (i.e., the planet Venus).

The fact that RDF collapses this distinction is not valuable from a 
technical standpoint, since the 'sense' of the meaning of the link is 
quite retrievable in XTM; it's the name of the tag the URI is attached 
to. And topic maps do not over complicate the issue; there are only 
three 'senses' that they utilize, and they provide a lot of utility. 
Saying essentially, "RDF doesn't make the distinction, therefore, RDF 
is good because it doesn't make the distinction," is blatantly begging 
the question.

Honestly, it bothers me that the "semantic web" would not bother to 
reflect such a simple philosophical concept in it's syntax. Have the 
folks at the W3C not realized that linguistic and philosophy experts 
have spent quite some time pondering the same problems that the 
semantic web seeks to solve?

Michael Della Bitta
Columbia University