- From: <Noah_Mendelsohn/CAM/Lotus@lotus.com>
- Date: Tue, 11 May 1999 19:07:02 GMT
- To: paul@prescod.net, simonstl@simonstl.com
- Cc: www-xml-schema-comments@w3.org
Of the schema structures editors, I am probably the least knowledgeable on
the nuances of entities, but let me give some preliminary perspective on
what you found in the draft specification. Our requirements document
(http://www.w3.org/TR/1999/NOTE-xml-schema-req-19990215) includes a design
principle stating that:
"The XML schema language shall be: ...more expressive than XML DTDs"
I think its fair to say that the editors took this as an instruction to
include features which, at minimum, model the capabilities of DTDs in
dealing with entities. The reason for the design principle, I believe, is
to ensure that a DTD can in general be converted to a schema, which no
ongoing need to retain the DTD.
It's my impression that at least some of the editors share the
reservations expressed about various aspects of the entity mechanisms, and
are thus reluctant to perpetuate them as we did in the new design. It
does appear that failing to do so would restrict one's ability to convert
arbitrary DTD's into equivalent schemas, and would entail a change of our
requirements document along with an associated change in the design. So
it's a tradeoff, and I don't think we've finally settled which way to go.
Your opinions are much appreciated.
I hope this helps shed some light on the reasoning behind the decision.
Also: keep in mind that the purpose of WD1 is in part to generate just
this sort of discussion. Our goal is not to defend this design to the
death but to learn more about the tradeoffs as we move forward.
Thanks very much.
Noah Mendelsohn
Lotus Development
Received on Tuesday, 11 May 1999 15:03:15 UTC