W3C home > Mailing lists > Public > www-rdf-comments@w3.org > October to December 2001

Re: Meaning of an RDF document: issue rdfms-assertion

From: Tim Berners-Lee <timbl@w3.org>
Date: Tue, 20 Nov 2001 09:38:43 -0500
Message-ID: <006801c171d1$0dc173d0$e9001d12@CREST>
To: "Dan Brickley" <danbri@w3.org>
Cc: "Aaron Swartz" <me@aaronsw.com>, "www-rdf-comments" <www-rdf-comments@w3.org>

----- Original Message -----
From: "Dan Brickley" <danbri@w3.org>
To: "Tim Berners-Lee" <timbl@w3.org>
Cc: "Aaron Swartz" <me@aaronsw.com>; "www-rdf-comments"
<www-rdf-comments@w3.org>
Sent: Monday, November 19, 2001 9:03 PM
Subject: Re: Meaning of an RDF document: issue rdfms-assertion


>
> So which predicates are semantically delgatory in nature? rdf:type, it
> seems? Any others? If this is a real distinction, we might consider asking
> WebOnt to create subclasses of rdf:Property for those that delegate to the
> subject, or to the object, of a statement...

rdf:type is special.  It is the hook for making unary predicates with a
mechanism
which natively provides binary predicates. Though anyone could define a
predicate which did the
same thing, or was for example the inverse of rdf:type, I don't sse any
reason
for making a special class of them.

> Not entirely convinced; this semantic atomism often feels like ascribing
> meaning to the letters that make up the words in a sentence,

Think of it as a question of hand-off between the specs.  This must be done
cleanly and is the basis of how basically all communication systems work.
The IP packet has a way of saying the packet is a TCP packet. The TCP
protocol has port number which is used, though the /etc/ports
registry, to point to a protocol specification for the protocol on top of
TCP.
HTTP has way (content-type) of saying what specification to use to
understand the body of the message, and if that is XML, XML has a way
(namespaces) to tell you which spec to read to understand what the document
means. Specifically, the key is the namespace of the outermost element
and its element name.  In the event that these are rdf:rdf, then RDF
defines (where many langauges do not)  how to extract each statement,
and oit must finish the job of defining which spec to look at to figure
out (or argue about) what the statement means.  The specification
is that of the predicate of the statement.  That's how it works.

The granularity is the URI.   No, it is not the letter.  But the whole
coolness of the semantic web is closely based on this fundamental
semantics at the low level.

> Dan
>
> On Mon, 19 Nov 2001, Tim Berners-Lee wrote:
>
> > Precicely.
> >
> > ----- Original Message -----
> > From: "Aaron Swartz" <me@aaronsw.com>
> > To: "Dan Brickley" <danbri@w3.org>
> > Cc: "www-rdf-comments" <www-rdf-comments@w3.org>; "Tim Berners-Lee"
> > <timbl@w3.org>
> > Sent: Monday, November 19, 2001 9:44 AM
> > Subject: Re: Meaning of an RDF document: issue rdfms-assertion
> >
> >
> > > On Friday, November 16, 2001, at 03:58  PM, Dan Brickley wrote:
> > >
> > > > Its the predicate that sets the meaning, unless the predicate
> > > > says that something else sets the meaning...
> > >
> > > I don't see how this is contradictory. If I say the Director is
> > > in charge, but he can delegate authority to a domain leader,
> > > does that lessen his authority? I'd say no.
> > >
> > > --
> > > [ "Aaron Swartz" ; <mailto:me@aaronsw.com> ;
<http://www.aaronsw.com/> ]
> > >
> >
>
Received on Tuesday, 20 November 2001 09:38:40 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:15:17 UTC