RE: universal languages

As I've been following all the stuff about URIs and global names, 
etc. I've been reminded of many years of fighting the issues of 
symbol grounding in my misguided youth (Pat, I can only claim 22 
years of AI, so you win)...
  However, I'm puzzled by Stefan's claims that we cannot deal with a 
lot of things without "true" URIs and some of the other things I've 
been reading here.

Here's a stab at something - to me the most important idea in DAML is 
the "equivalento" sort of mechanism -- it let's a first user define a 
desiognator for Boston, a second user defines a different URI for 
Boston, and a third comes along and says - they are the same (the 
issue of whether to believe the third person is irrelevant to this 
discussion - all multi-person representation will need some sort of 
caveat emptor mechanism or we'll all be in infinite loops forever)

So, I say on one page that (forgive me for not using RDF, DAML or the 
like - this doesn't need the syntactic stuff)

  jah:ProfessorHendler
is at the URI
  http://www.cs.umd.edu/~hendler/#hendler-daml

Someone else has that
  darpa:JimHendler
is at URI
  http:"mailto:jhendler@darpa.mil"

If I now put somewhere in web space that
  EquivalentTo jah:ProfessorHendler darpa:JimHendler

then we have a nice network, if you believe me we've appropriately 
grounded the symbols, etc.

btw, you're free to have your own completely different notion of what 
a James Hendler is, but if you want to share it with me, you'll need 
to "name" it, and then I'll need to be able to find it on the web, 
and also to check my KBs to see if I know something it is 
equivalentTo.

  Hey, I like the web -- let's use it!

  -Jim H.  (URI = PGP-key:123451234543212345432...)
Prof. James Hendler		hendler@cs.umd.edu
Computer Science Dept		703-696-2238 (phone)
Univ of Maryland		703-696-2201 (Fax)
College Park, MD 20853		http://www.cs.umd.edu/~hendler

Received on Tuesday, 6 February 2001 14:17:43 UTC