RE: URIs: resources and contradictions was: Re: httpRange proposed text

> What is left ungrounded then is the meaning of a URI. This can be
> grounded either by a model theoretic interpretation (RDF), the history
> of representations (REST), or some other technique.

Exactly!!!

I think of it this way:  

* How can RDF be used to ground the meaning of a URI without relying on at least *some* grounded URIs to begin with?

* How can the history of representations be used to ground a URI if there is no guarantee that the representations will come from the same "thing" on successive GETs?

If you try to ground the meaning of a URI inside the system, the system breaks down.  That is why the axiom of identity says nothing about *how* an identifier gets its meaning, it simply asserts that an identifier *does* have meaning.  

And if you don't assert *that*, you can't ground *anything*, let alone URIs.  That's why this axiom is the most fundamental axiom.

And FWIW, I think Roy even agrees with me about *how* URIs become associated with meaning.  "URIs mean what people use them to mean".  The fundamental "anchors" that ground everything else are based on common usage.

That's exactly the way that words work as well.  "Words mean what people use them to mean", and common usage changes over time.  Common usage can be influenced by organizations such as the French Academy of Language, merging cultures of common usage (e.g Spanish and English), and many other ways. 

This fluidity and sometimes ambiguity does not reduce the importance of words in anchoring our real-world semantic communications.  Philosophers complain mightily that "it is impossible to guarantee that words have the meaning you intend", but it is really irrelevant when it comes time to communicate semantics between ourselves, because words are all we've got.  If we didn't have words, we would invent them.

This is, in fact, the reason that Dag Hammarskjold's admonition is so important.  He realized that words can be used ambiguously, and can even be used in ways that deliberately distort semantics.  He recognized that the ability of words to unambiguously identify is not guaranteed by the nature of words themselves, but by the way that we USE words.  Or to use your terminology, words do not ground themselves -- words are the things that WE have to ground if we ever hope to ground anything else.  He says "respect for the word is the first commandment".

Received on Sunday, 4 August 2002 14:54:25 UTC