W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > August 2003

Re: RDF Use Case: scraping metadata from the web

From: Martin Duerst <duerst@w3.org>
Date: Sat, 02 Aug 2003 11:30:13 -0400
Message-Id: <4.2.0.58.J.20030802104033.06a2e0f8@localhost>
To: Brian McBride <bwm@hplb.hpl.hp.com>
Cc: rdf core <w3c-rdfcore-wg@w3.org>, i18n <w3c-i18n-ig@w3.org>

At 22:49 03/07/31 +0100, Brian McBride wrote:

>Martin Duerst wrote:
>
>[...]
>
>>
>>We can always record any kind of information in the graph. But the
>>problem there is that different people/applications will do this
>>differently.
>
>I don't think that's a problem in this use case, as it is defined.   Do 
>you want to have a go at refining the use case to illustrate the problem 
>you see?

Hello Brian,

I think this discussion about use cases is very useful,
and is helping us a lot advancing our common understanding.

Rather than refining the use case, let me extend it. Assume that
the RDF created by scraping all kinds of <title>-like elements from
all kinds of document formats is viewed in some kind of RDF tool.
This tool may use language information to disambiguate glyph shapes,
or it may use language information to appropriately speak the texts
to users with disabilities. If there is a clearly defined and
uniform way to have language information for XML literals, then
that can easily be done. Otherwise, this informations is most
probably just ignored.

Please note that for my extension of the use case, it would be
enough to have xml:lang on a <dummy> element internal to the
XML Literal. But that, as we have discussed, does not work for
the original part of the use case, where we were concerned with
markup integrity. We can solve half of the use case with one
solution, and the other half with the other solution, but the
solutions are different and so don't work together.


Regards,  Martin.
Received on Saturday, 2 August 2003 11:43:48 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:59:32 EDT