W3C home > Mailing lists > Public > www-rdf-logic@w3.org > February 2001

Re: universal languages

From: Sean B. Palmer <sean@mysterylights.com>
Date: Tue, 6 Feb 2001 21:08:10 -0000
Message-ID: <005e01c09081$110cb400$f845073e@z5n9x1>
To: "pat hayes" <phayes@ai.uwf.edu>, "Stefan Decker" <stefan@db.stanford.edu>
Cc: <www-rdf-logic@w3.org>
> (a) Global unique object identifiers are necessary for the
>       context-free representation and exchange of data on the web.

As Dan pointed out, I'm pretty sure that we *all* agree that some kind of
representation of objects in needed for a hypermedia environment. Still,
what Pat has been debating is that defining things outside of this
hypermedia environment is therefore impossible. Being an Astrophysicist, I
like to compare that to talking about things that are outside the universe.
There is no outside the universe, because the concept of space is only
something that can exist throughout the universe...

However, we can conceptualize to bring these extranneous objects into our
environment, and give them some meaning (as Sandro points out). I can talk
about something, and use that as a proprietary representation, but we must
not confuse that with a definitive representation. What this all boils down
to is trust: if I trust a fundamental principle of a "zip code" and someone
else trusts it as well, and understands the same contexts set up by that
property, then we can communicate our data.

Some of the URI confusions are simple name vs. address debates. This is
both a fundamental flaw of the Semantic Web, and at the same time the thing
that will enable it to have a tremendous growth capability. I am sure that
although dereferencing resources may seem propretary, it is a neat way to
define something, rather than have a processor guess it. This is why RDDL
is taking off at such a tremendous rate. To use one of Aarons expressions:-

     "If a namespace falls in a forest and there is no schema around
      to define it, does it make a sound?"

This was in response to my pointing out that a namespace is not
definitive... but it's all about trust, that's what separates the triples
model (which Pat points out we are "revereing" to vehemently) into a neat
machine processable structure for the interchange of data.

--
Kindest Regards,
Sean B. Palmer
@prefix : <http://webns.net/roughterms/> .
[ :name "Sean B. Palmer" ] :hasHomepage <http://infomesh.net/sbp/> .
Received on Tuesday, 6 February 2001 16:10:08 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:52:38 GMT