W3C home > Mailing lists > Public > www-archive@w3.org > March 2004

Re: more text

From: Chris Bizer <chris@bizer.de>
Date: Tue, 16 Mar 2004 12:34:19 +0100
Message-ID: <006101c40b4a$a25e21e0$1f12fea9@named4gc1asnuj>
To: "Jeremy Carroll" <jjc@hpl.hp.com>, <patrick.stickler@nokia.com>, <phayes@ihmc.us>
Cc: <www-archive@w3.org>

A few first comments on the structure of the document:



I would propose to separate two things:

1. First we define the concept of named graphs, a general extension to RDF
which can be used in a wide range of scenarios (many of which we clearly
cannot image now).

2. Then we describe two *use cases* of named graphs: First our agent
scenario, which is essential for the semantic web as a distributed system,
but still is only one possible *use case* of named graphs. Second, the
combination of named graphs with logical vocabulary, which is also a use
case.



I was talking to a guy last weak, who was using RDF to model water in order
to make flood forecasts (?!?). He was highly interested in named graphs for
modelling some context related aspects in his model, but wasn't interested
in our agent/asserting/affirming stuff at all. Thus I think it is essential,
that we define named graphs in general manner, open for different use cases
(chapter 2,3,4 and 5) and present our thoughts about provenance and trust as
one use case of named graphs (chapter 6 and 7) and our thought about
paradoxes as a second use case (chapter 8).



Thus I would also mandate two different namespaces: rdfg: for general terms
like Graph, subGraphOf, equivialentGraph ... and swp: (short for: Semantic
Web Publishing) for our agent scenario specific vocabulary.



It might also be a good idea to restructure chapter 6 and 7 corresponding to
the roles of the agents involved in the scenario.



Role1: Information Provider (the guy who asserts or quotes stuff)

Role2: Information Publisher (the guy who makes graphs accessible)

Role3: Infomration Consumer (the guy who uses the information and descides
which information he trusts)



Maybe role 1 and 2 don't have to be separated. I will think further about
this.



His would mean that we put everything about role1 and 2 (asserting and
publishing including the signing stuff) in chapter 6-Provenance. We might
also want to rename the chapter to "Publishing information on the Semantic
Web". And put everything related to role 3 in chapter 7 -Trust (policies,
signeture verification), which could be renamed as "Using information found
on the Semantic Web". (Renaming optional :-)



Comments on the vocabulary section will follow .....



Chris
Received on Tuesday, 16 March 2004 07:32:44 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 22:32:25 UTC