Re: Valid XHTML and Faithful Infoset redflag - stated (semi) formally

On Wed, 2007-04-25 at 10:26 -0400, Chimezie Ogbuji wrote:
[...]
> Define a class xhtml:Document defined (necessarily and sufficiently) as
> the class of documents which have a jc:validWRT relationship between
> them and a W3C sanctioned XHTML DTD

That excludes the input document in the test case currently in question.
http://www.w3.org/2001/sw/grddl-wg/td/html-and-grddl-xform-attr

jjc objects to that objection, with good reason.
http://lists.w3.org/Archives/Public/public-grddl-wg/2007Apr/0129.html

And actually, DTDs funamentally don't fit the jc:validWRT model.
xml:valid(doc) is a one-place predicate, and it's only
true when doc includes (by value or by reference) a DTD.
It would be really nice if XML 1.0 had defined
a valid(doc, DTD) relationship, but it didn't.

-- 
Dan Connolly, W3C http://www.w3.org/People/Connolly/
D3C2 887B 0F92 6005 C541  0875 0F91 96DE 6E52 C29E

Received on Wednesday, 25 April 2007 14:53:00 UTC