Content or Metadata?

[Rick Jelliffe]

From: Robin Cover <robin at isogen.com>

>Of course, such notions reflect perspective, which may or may
>not be implicit/explicit in the style rules and underlying
>assumptions of the house.

For all its sins, RDF showed up a major area that is currently missing
in Schemas: the need to make the generic relationships between elements
explicit.

In particular RDF used "bag", "seq" and "alt".  But there are many more
such relationships:
    * is one element an annotation of another?

    * is that annotation superior (e.g. a title, a summary) or
subsidiary (e.g., an explaination, a digression, an alternative, a
role)?

    * does one element/attribute have any meaning without some other
element/attribute (e.g., does a particular number also require a units
element/attribute/default)?

    * which roles do elements and atributes play in the particular
taxonomic/ontological methodology of their creator (e.g., what is data,
what is metadata)?

Some of these things, RDF Schemas could make possible, and XLink could
have made possible.  I think RDF is a continual reminder that GIs and
containment may make relationships obvious to humans, but in the absense
of other conventions, they may hide these relationships from the
computer.

B.t.w, the sins of RDF were all commented on at the time:

  * the spec is clearly two or three different different
documents cobbled together with little cohesion between them;

  * having a syntax like the _n attribute names which made validation
impossible except by special-purpose validators;

   * not having the discipline of a DTD fragment, so that some elements
mentioned are never explicitly given in the EBNF productions;

   * RDF is a framework but it should have been an architecture which is
framework-neutral. The test of whether it is useful as a framework are
whether generic tools are useful for RDF data; if, in fact, it is being
mainly used for specific applications, then RDF markup would be better
formulated as conventions that sit on top of DTDs/schemas that allow as
natural modeling of the data as possible.

I think RDF should have concentrated on how to fit on top of
regular markup, including markup of inline elements interspersed
through paragraphs. Atomic data is just the simplest case of that.


Rick Jelliffe



xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at 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 at ic.ac.uk the following message;
unsubscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)