Do you or Dont you buy Tim Bray's Namespace Validation Algorithm?

Mark Tucker mct at
Thu Sep 10 22:09:25 BST 1998

Wait, People, 

	DTD validation in the face of Namespaces:
	    "Re-write the document instance with consistent Prefixes
	     then do a normal DTD validatation."

	I don't see anything kludgy in it. (modulo my preference to
        use expanded names directly in the processor's symbol table.)

	It just says: 

		1. Determine the Expanded the names the way the
		Namespace Proposal says. (Section 6.3)
		2. Define a unique prefix for each namespace definition URI .

		3. Rewrite the element or attribute, prepending the
		(possibly generated) unique prefix for the namespace
		of the element/attributes Expanded Name.
		   This give you  P1:BOOK, P1:NAME, P3:ADDRESS
		   if BOOK and NAME come from the same namespace URI.
		4. Do the same to the DTD's that you read in.
		5. Do normal DTD validation of the rewritten instance 
		   document against the rewritten DTD.

It seems to me that DTD's with namespaces CAN be validated.

"Richard L. Goerwitz" <richard at> wrote

rg> Then they won't validate.  That's all.  Or (lacking an associated
rg> DTD), they could be validated in the kludgy fashion that the current
rg> namespace spec seems to require (i.e., you have to make your DTD aware
rg> of all the new:elements your use of namespaces requires - with
rg> allowances for de- faulting :-( ).

	You do not have to rewrite your DTD at all.  You  do not have
	to "merge content models", and "introduce ANY".

	The whole validation process is mechanical, and gives the
	expected results.
Mark Tucker			tucker_m at
Regenstrief Institute		phone: (317) 630-2606
1001 W. 10'th St; Indianapolis, IN; 46202-2859;	fax: (317) 630-6962

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at
Archived as:
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