W3C home > Mailing lists > Public > www-rdf-interest@w3.org > September 2001

Re: Understanding RDF as extensible vs. embeddable

From: Brian McBride <bwm@hplb.hpl.hp.com>
Date: Sun, 30 Sep 2001 09:06:48 +0100
Message-ID: <3BB6D298.80901@hplb.hpl.hp.com>
To: Jon Garfunkel <jgarfunk@coopdata.net>
CC: www-rdf-interest@w3.org

Jon Garfunkel wrote:

> Hello RDF developers, I would appreciate your help on understanding a basic
> aspect of RDF which I've been trouble finding an explanation on.

I've copied this response to the rdf interest mailing list which is
probably a better forum for this sort of question.  You may get further
responses there.

> I have a
> fair amount of experience with XML/XSLT, and am now having a look at RDF. I
> explain more about the usage below. I'll call the schema/top-level element
> I'm working with "E". I am wondering how to mix it with RDF. Two possible
> ways:
> By "extensible" I am thinking of object inheritance (E "extends" RDF)
> and by "embeddable" I am thinking of a variable (E "has variable" RDF)
> In the HTML examples for RDF, the model is necessarily embeddable: we deal
> only with HTML documents, so the RDF information appears in embedded META
> tags.
> In all the XML examples I see, the XML documents appear to have their E
> information embedded within an RDF structure. I understand that this is
> *functionally* extensible, and thus a good strategy. And it of course is a
> helpful high-level interface.
> Still, I am looking to embed the RDF metadata in my E documents. Does this
> have any present or future 

Yes.  It is intended that RDF can be embedded in XML documents.  I don't
have much experience of how present tools cope with this structure.  I'm
not sure which parsers can be set to pick out just the RDF in an XML
document, or give the application control of which elements to process as
RDF.  Perhaps your XSLT skills can be used to isolate the RDF parts of
your documents for parsing.

The standard way to identify blocks of RDF is to include them in
<rdf:RDF> ... </rdf:RDF> brackets, though M&S does not require this.
The working group has still to discuss some issues around handling
multiple such elements in a single document.

>(I use several different schemas) I can give a
> few reasons:
> 1. Conceptually, I like to think of my E's as the specific things they are
> (like "entity", "presentation", "agenda"), instead of as a "RDF Resource
> which contains an 'entity'". But I'm open to conversion.
> 2. I need to rewrite some DOM, XSLT, XML code to reflect this. Again, this
> is a one-off effort.
> 3. I'm very satisfied with my XML editor, XMLSpy 3.5. But neither it nor
> version 4.0 seem to enable the creation of RDF-based XML schemas out of the
> box (unless I'm mistaken). So instead it seems easier to embed my own pidgin
> metadata.
> 4. I don't even know how to create a schema for E which embeds/extends RDF
> (or visa-versa). I've been pleased with using the XML Schema Primer for
> reference, but I haven't found any examples for creating E extending RDF.
> Further complicating this is that I get the sense that RDF schemas are
> orthogonal to XML schemas.
> The overall purpose is a web-basd collaboration system. It has similar
> functions to the Annotea project (I've been following their updates on the
> www-annotation list) among other capabilities. I'm focusing on a
> server-based solution, so I have a module which reads all of the XHTML, XML
> documents on a designated area of the filesystem. Thus I am looking for a
> solution to get the metadata from XML documents.
> Thanks,
> Jon

Received on Sunday, 30 September 2001 04:11:16 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 22:44:32 UTC