W3C home > Mailing lists > Public > www-rdf-interest@w3.org > March 2005

Re: SemWeb Non-Starter -- Distributed URI Discovery

From: Danny Ayers <danny.ayers@gmail.com>
Date: Sat, 19 Mar 2005 12:21:47 +0100
Message-ID: <1f2ed5cd05031903215e24ea59@mail.gmail.com>
To: Stephen Rhoads <rhoadsnyc@mac.com>
Cc: semantic-web@w3.org, www-rdf-interest@w3.org

On Fri, 18 Mar 2005 17:49:39 -0500, Stephen Rhoads <rhoadsnyc@mac.com> wrote:

> As far as I can tell, there is no formal, generalized mechanism to reliably query the owner of a URI in order to obtain an RDF Description of that URI.  And this is a serious impediment to the Semantic Web.

Yep, and arguably it extends beyond just query -

>From URIQA [1]:
[[
As the Semantic Web emerges and the behavior of automated software
becomes increasingly directed by explicit knowledge about resources,
gathered from disparate sources, the need for a standardized means of
sharing authoritative knowledge about a given resource, based solely
on the URI denoting that resource, becomes critical to achieving a
fully open, global, scalable, and ubiquitous Semantic Web.
]]

I agree this is an important point, as the Web* is built on HTTP+URIs
and the Semantic Web seems so far only to have got to URIs.

There have been efforts such as RDDL [2] to cover the query/GET side
in a manner that would be suitable for both the current Web and the
Semantic Web. As Joshua suggests, it's probably a bit late to earmark
slashes or hashes for the purpose. But is the query side enough?

I'm personally not comfortable with URIQA's approach, addition of new
HTTP verbs - it's a nice technical solution, but I'd worry about
deployment. But maybe we do need something you could loosely describe
as the RDF Protocol for description-oriented actions (covering the
ASK, TELL kind of agent language, as expressed in HTTP as GET, PUT
etc).

But if the query side is enough then presumably (as you suggest) the
solution may well lie in the SPARQL protocol.

Cheers,
Danny.

*Ok, as danbri has been saying on the Atom list, the "Web" as we know
it can be characterized in other ways, and isn't necessarily
HTTP-specific. But as the primary protocol at this point in time, it
should probably be taken into consideration...

[1] http://sw.nokia.com/uriqa/URIQA.html
[2] http://www.rddl.org/

-- 

http://dannyayers.com
Received on Saturday, 19 March 2005 11:21:48 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:52:13 GMT