FYI--off-topic--Notation Schemas proposal

Rick Jelliffe ricko at
Thu May 13 19:00:00 BST 1999

For anyone who is interested, there is a draft-in-progress of my ideas
for a good schema language.

It allows you do things like

* content models for attribute lists: to say element X can have
  (id & (size & size-units)? & ( href | entityref)? )

* content models for attribute values, to say that attribute Y can have
     (SATURDAY? & SUNDAY?) )

* content models for IDREFS links, to say that element Z can have IDREFS
  particular types of elements
  ( ( p | q | tbl ), (p, | q | tbl)

In other words, it treats attribute lists, attribute value lists, &
IDREF lists with the
same strong typing that is available to elements.  The foundation of my
is that everything should be modelled as a grammar: if the decision of
to make something an attribute, an element or an element reference
(IDREF) is
completely arbitrary (from the information modeled), then why not allow
strong typing on all them.  It moves XML from being a strongly-yped tree
language to being a strongly-typed graph language.

It also allows people more freedom to structure their documents, for
example to
move to flat relational-style tables without giving up strong typing.
This would
make RDF more useful too.

However, I gather some of this might be out of the scope of the XML
Schema group.
However, since apparantly XML 1.0 conformance was outside their scope
apparantly (I am being naughty: they did an amazing job to pull together
many proposals and put it out on time...lets not forget that it is only
a draft),
perhaps they may be open to developing in other directions.

Rick Jelliffe

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at
Archived as: and on CD-ROM/ISBN 981-02-3594-1
To (un)subscribe, mailto:majordomo at the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at

More information about the Xml-dev mailing list