W3C home > Mailing lists > Public > public-sw-meaning@w3.org > November 2003

Re: Terms and statements (was: consensus and ownership)

From: Graham Klyne <gk@ninebynine.org>
Date: Thu, 06 Nov 2003 13:46:06 +0000
Message-Id: <>
To: Dan Connolly <connolly@w3.org>
Cc: public-sw-meaning@w3.org

At 12:02 17/10/03 -0400, Dan Connolly wrote:
>On Thu, 2003-10-16 at 09:52, Graham Klyne wrote:
> > At 08:47 16/10/03 -0400, Thomas B. Passin wrote:
> > >Graham Klyne wrote:
> > >>This idea of meaning being based in consensus also appears in the 
> work by
> > >>Quine that I mentioned the other week [1].
> > >>A possible difference in position would be that you talk about the
> > >>meaning of a URI, where Quine's analysis suggest that it's not the
> > >>individual terms but complete statements that have meaning.  (I think
> > >>that's a point that Pat has been trying to press, too.)
>Yes, but an interesting thing about the Web is that many of the
>terms refer to documents, i.e. collections of statements.
>And if you know what <http://domain/doc> means, then
>it gives you a pretty good handle on what <http://domain/doc#term>

That a term refers to a document doesn't seem, to me, to really tell you 
what it means in any definitive sense.  The remainder of your message seems 
to reinforce this uncertainty.

One thing we can do about URIs that reference documents is make statements 
about them that the web community would generally hold to be true; e.g.

    HTTP-GET(some-uri,other parameters) == (some bag-of-bits)

In another message, you said:
Anyway... I think the general case is that a
URI has meaning to the extent that there's consensus
in the Internet Community about what it means, as expressed
in Internet protocol messages, especially messages that
express a relationship between a URI and a representation
of what it means; and that the HTTP/DNS case is, while very
common, a special case where the Web Community has delegated authority
to one party
-- http://lists.w3.org/Archives/Public/public-sw-meaning/2003Oct/0084.html

I's find it easier to agree if this said:
"... to the extent that there's consensus in the Internet Community about 
what statements containing it are true ..."
and a feature of the qualifications you add, I think, would be that 
corresponding statements are testable in a Web environment.


>My thinking on this is very much in progress... there are
>complications like
>   - In general, URIs refer to stateful resources, not
>     documents in the sense of sequence-of-characters.
>     But... is there a useful idealization that hides this wrinkle?
>   - how about looking at resources as agents; i.e. GET
>     is sorta "tell me (or give me a form to inquire about)
>     everything you know, to the extent you can in format XYZ".
>     So foo#bar is what foo means by bar.
>     I spend a certain amount of time swimming around the literature
>     of logics of knowledge, modal logics, etc.
>Dan Connolly, W3C http://www.w3.org/People/Connolly/

Graham Klyne
For email:
Received on Thursday, 6 November 2003 09:12:34 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 19:56:01 UTC