xml-dev - RE: RDF, again

[Vane Lashua <vlashua@RSGsystems.com>]

Asking out of ignorance: 
Is there thought being devoted to a universally accessible catalog of id's,
names (lists), classes, datatypes -- maybe even using the MARC system and
the LC index -- existing as a universal repository of components describing
data structures?

It would be a "soft" resource, like a library catalog, but with "hard" data
points: the LC system is not a standard; it is a registry and a reference
maintained by an authority. A publisher may suggest the cataloguing
classification of an individual object, but any given library may catalog
its instance-object differently. Meanwhile, because publishers and libraries
are interested in keeping in touch with information, a library patron from
virtually anywhere can find most objects in a given class and select from
them.

The difficulty with the definitions below, for instance, is that "name" is a
collection of characters whose context is not clear without a reference.
Namespaces, it seems to me, are absolutely necessary, but they tend to
encourage diversity where convergence would be a more enlightened tendency.

Vane


-----Original Message-----
From: Mark Birbeck [mailto:Mark.Birbeck@iedigital.net]
Sent: Tuesday, November 23, 1999 6:46 PM
To: 'Paul Prescod'; 'xml-dev@ic.ac.uk'
Subject: RE: RDF, again


Paul Prescod wrote:
> The thing I find confusing about the RDF syntax is that the 
> element type
> name can be either an RDF type name or an RDF property. XML makes no
> distinction and that's why I think that it is difficult to use for
> object oriented interchange.

I got the impression from the spec that this is intentional, so that a
straightforward XML document - that might not contain *any* RDF - can
still be interpreted as a set of RDF statements. In other words,
different XML layouts (elements for attributes, e.g.) of the same data
would result in the same RDF statements.

The XML would still need to be well thought out though. For example:

	<person name="Paul">
		<food>trifle</food>
	</person>

might mean trifle is your favourite food, the main food you're allergic
to, or your pudding preference for the office Xmas party. All of these
are acceptable in XML, but the RDF interpretation of this may well be
incorrect - or at least not as rich in meaning as we would like:

	Person has a name "Paul" and a food "trifle"

So, to make the first statement - trifle is Paul's favourite food - we
could use the following RDF:

<rdf:RDF>
	<rdf:Description ID="1">
		<rdf:Type rdf:resource="person" />
		<x:name>Paul</x:name>
	</rdf:Description>

	<rdf:Description ID="2">
		<rdf:Type rdf:resource="food" />
		<x:name>trifle</x:name>
	</rdf:Description>

	<rdf:Description about="#1">
		<x:favourite rdf:resource="#2" />
	</rdf:Description>
</rdf:RDF>

Using the abbreviated forms allowed to us, this is the same 'RDF':

	<x:person x:name="Paul">
		<x:favourite>
			<x:food>trifle</food>
		</x:favourite>
	</x:person>

or:

	<x:person>
		<x:name>Paul</x:name>
		<x:favourite>
			<x:food>trifle</x:food>
		</x:favourite>
	</x:person>

or:

	<x:person>
		<x:name>Paul</x:name>
		<x:favourite x:food="trifle" />
	</x:person>

So - to turn this round - any of the previous three XML documents can be
interpreted as the same set of RDF statements - a person with the name
"Paul" has a favourite food, and that food is called trifle - even
without any explicit RDF present.

As to whether it is any good for object interchange, I think it is. Of
course, if the relationships between elements contained within other
elements can be inferred then straight XML is fine. But as soon as you
need something more complex then RDF is very good (not to mention when
the objects being referred to are outside of the XML document you're,
and so you can't use ID/IDREF.)

Best regards,

Mark

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev@ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN
981-02-3594-1
To unsubscribe, mailto:majordomo@ic.ac.uk the following message;
unsubscribe xml-dev
To subscribe to the digests, mailto:majordomo@ic.ac.uk the following
message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa@ic.ac.uk)

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev@ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To unsubscribe, mailto:majordomo@ic.ac.uk the following message;
unsubscribe xml-dev
To subscribe to the digests, mailto:majordomo@ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa@ic.ac.uk)