RE: DC in XHTML2

Dan wrote:

> I'd be interested to learn more about news organization 
> requirements/aspirations regarding the structure of the things a 
> dc:subject might point to, and in particular on getting review 
> comments on the SKOS working drafts there...

With the conclusion of IPTC's 40th AGM today, I hope to have more 
time to participate in this thread.  

For some background, our timescales and some fantasy XML, see the 
mail below.  We need help to translate our requirements into XHTML2 
and Semantic Web constructs.  Where there is a mismatch, we wish to 
negotiate and, hopefully, reach a mutually acceptable solution.  We 
are interested in SKOS, but haven't had the time to do more than 
glance at it.

What we would like is some intensive collaborative discussion, 
preferably a face-to-face meeting between the appropriate members of 
the IPTC News Architecture WP and the appropriate people from the 
RDF in XHTML Taskforce.

Misha


-----Original Message-----
From: Misha Wolf 
Sent: 26 May 2005 21:33
[snip]
Subject: RE: Info on news standards vs. Semantic Web?

[snip]

Many thanks for your contribution to the News Standards Summit.  In 
response to your note (below), I'll summarise the current situation 
and make a proposal.

I can speak only of the mainstream news industry (news agencies, 
newspapers, broadcast media, large online news sites), not for the 
blogging (Atom/RSS) community.  The news industry develops its 
standards within the International Press Telecommunications Council 
(IPTC):
   http://www.iptc.org/

The handling of metadata in our existing standards is based on a 
TopicMaps intercept called TopicSets, designed about five years ago. 
There is no use of URIs to identify vocabularies or terms, and the 
vocabularies are fairly primitive: either flat or with a hardwired 
number of levels.

In the context of a program of work called IPTC Roadmap 2005, the 
members of the IPTC are building a completely new news architecture, 
to be based, where possible, on the latest applicable standards.  I 
chair the News Metadata Framework WG, which has been tasked with 
defining how metadata will be handled in this new architecture.

We have decided that all terms will be located within vocabularies 
identified by URIs and that the terms themselves will be identified 
by URIs.  We have not chosen a syntax as yet, but have agreed on 
various requirements (in particular relating to bandwidth) which 
would prohibit the use of the original RDF serialisation.

I should explain that a typical headline broadcast by Reuters has 50 
or so associated codes.  If the headline needs, say, 30 chars for 
the text, we're not willing to spend, say, 10,000 chars on the 
metadata.

We have tentatively decided to use QNames in attribute values, eg:

   <code
      val="scheme:code"          <!-- Value -->
      type="scheme:code"         <!-- Concept type -->
      qual="scheme:code"         <!-- Qualifies -->
      parent="scheme:code"       <!-- A parent -->
      why="scheme:code"          <!-- Why present -->
      equiv="scheme:code"        <!-- Equivalent to -->
      assign="scheme:code"       <!-- Assignee -->
      conf="50"                  <!-- Confidence (%) -->
      rel="50"                   <!-- Relevance (%) -->
      when="2005-05-11T12:34:56" <!-- Date and time -->
      xml:lang="en"              <!-- Refers to the Description -->
   >Hello world</code>           <!-- Description -->

and have been pondering how to relate this to RDF.  We were, 
therefore, intrigued by Steven's talk at the recent News Standards 
Summit, in which he told us how the HTML WG is thinking of solving 
this problem for XHTML 2.0.

Our plan is:

   6 June - 9 June  : IPTC AGM, where the broad outline of the new 
                      architecture will be reviewed and the next 
                      steps agreed on

   June - September : All details (incl syntax) resolved

   24-26 October    : The architecture specifications agreed at the 
                      autumn meeting of the IPTC

We are very keen to work with the W3C to ensure that the News 
Metadata Framework fits into the Semantic Web.  We wonder whether 
it would be possible to set up some liaison forum where we could 
work, rapidly, over the next few months, to agree on an approach 
which fits both the needs of the news industry and the ideas 
underlying the Semantic Web.

For a recent version of the News Metadata Framework, see the IPTC 
home page:
   http://www.iptc.org/

News Standards Summit 2005:
   http://www.newssummit.org/2005/


Misha Wolf
Standards Manager
Reuters




--------- -------------------------------------------------------
        Visit our Internet site at http://www.reuters.com

To find out more about Reuters Products and Services visit http://www.reuters.com/productinfo 

Any views expressed in this message are those of  the  individual
sender,  except  where  the sender specifically states them to be
the views of Reuters Ltd.

Received on Thursday, 9 June 2005 18:06:29 UTC