W3C home > Mailing lists > Public > public-rdf-dawg@w3.org > January to March 2004

Re: Discover and publish a server's functionality

From: Kendall Clark <kendall@monkeyfist.com>
Date: Fri, 19 Mar 2004 09:06:51 -0500
To: public-rdf-dawg@w3.org
Message-ID: <20040319140651.GB28918@monkeyfist.com>

On Fri, Mar 19, 2004 at 09:44:04AM +0200, Patrick Stickler wrote:
> 
> 
> This is a nice specialization of the more generically stated use cases
> 
> PS-1: Discovery of authoritative knowledge via URI
> PS-3: Third-Party Knowledge Discovery; query by known identity
>
> which apply to any resource whatsoever denoted by a URI, such as
> a web service.

I think so, yes; though I reserve the right to change my mind after
actually thinking about it. :>
 
> Thus, if the client software knows the URI of the web service in
> question, it can then (hopefully) use the DAWG solution to obtain
> a description of that web service, and decide how best (or if) to
> interact with that service.

Yes.

> "Negotiation" is taken to equate to the client learning what
> protocols, query languages, parameters, etc. the server supports
> and deciding for itself how best (or if) to proceed.

Exactly; and, though I guess this is already clear, I want clients to
be able to "learn" these things via RDF. (This is why I'm trying to
analogize to con-neg, though this is really more like capability
negotiation.)

The only diff in my use case and yr 1 & 3 is that I think DAWG will
have to -- and this takes us beyond *use cases* strictly speaking --
work out some actual RDF vocabulary and conventions. (I should look at
the capabilities stuff the W3 has already done, of course; but first I
wanna finish writing more UCs.)

Kendall
Received on Friday, 19 March 2004 09:08:26 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 16:15:18 GMT