RE: Documents, Cars, Hills, and Valleys

> > > Here's another analogy. If I show you a photograph of the Eiffel
> > > Tower and ask you "What's that?", then I think that either of
> > > these two answers would be acceptable,
> > > * It's the Eiffel Tower.
> > > * It's a photograph of the Eiffel Tower.
> >
> > This example doesn't apply.  A URI is an *identifier*, a much more
> > appropriate comparison would be with "a word".
> 
> Which is exactly what I said.
> The identifier here is "that" (supplemented with a gesture).
> 

Nope, this is why I keep rejecting your position.  URIs are intended to
be first-class identifiers.  URIs are the same as words.  To say that
all URIs have the same level as ambiguity as the word "that" is to say
that all *words* have the ambiguity of the word "that".

It's not a matter of right or wrong -- if I accepted your position that
URIs are so ambiguous, and accepted that you need to disambiguate *all*
URIs, then I could attack your disambiguation scheme the same way that
you are attacking URIs.  It is possible to create ambiguity in anything.
The issue is pragmatism.  If every conversation required everyone to
disclaim every identifier, nobody would be able to communicate.  The
waves of word connotations that we surf are entirely capricious and
arbitrary.  Pointing out that they *are* arbitrary does not diminish
their usefulness, and normally only brings pleasure to people like Noam
Chomsky and Umberto Eco.

Received on Monday, 22 April 2002 18:12:09 UTC