-----Original Message-----
From: John F. Sowa [mailto:sowa@bestweb.net]
Sent: Thursday, March 04, 2004 7:58 PM
To: Richard Ballard
Cc: portal; John C. Mallery; BCNGroup Board
Subject: Re: a "redacted" public document
Dick,
I fully agree that there are serious problems in
natural language understanding. But those problems
are not the result of NL syntax, but of the semantics
of the subject matter.
I completely disagree with the following statement:
> ... The semantic web is the unambiguous representation
> we must jump to first, from there international natural
> language ambiguities can be (better) resolved subsequently
> with "forward semantics to linguistic processes."
That statement is a paraphrase of what Frege and Russell
said about predicate calculus over a century ago.
The only significant difference between predicate calculus
and XML is that XML is far more verbose. If philosophers,
linguists, and computer scientists have not solved the
problems of semantics by using predicate calculus as their
semantic representation, programmers and businessmen aren't
going to solve them by using XML.
John