Andy wrote:
> What if
> there are zero DC properties?
>
Well: all (!) DC stuff is optional and repeatable.
The record, which makes NO assertions is a simple DC record.
<?xml version="1.0"?>
<dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://purl.org/dc/elements/1.1/
http://www.dublincore.org/schemas/xmls/qdc/2003/04/02/dc.xsd"/>
appears as valid.
Obs!
The term "Namespace" doesn't give us anything. It does not help to identify the
vocabulary used. It does not help to determine a modelling style.
In particular in an XML Schema binding, another XML schema could come up
with different elements/attributes/types
for the same namespace - and that IS what happens.
How a given encoding is to be interpreted by means of an abstract model
IS a problem:
We have the good old HTML encoding for DC(Qual): It represents properties NOT
as XML elements or XML attributes, but as attribute values.
DC in HTML used dc:identifier to supply a referent which was supposed to
distribute. [This (together with the restrictions to values with the content model for HTML
meta's content attribute)
caused the well known 1:1 problem, which gave us so much headache.
But suppose you create an XML Schema which is a bit more flexible, then these issues
twindle - A different story].
I caution NOT to confuse
properties with elements or attributes in the sense of XML.
A similar issue is with nesting in XML -
....and XML namespace...
rs
|