W3C home > Mailing lists > Public > www-tag@w3.org > June 2007

Re: using DTDs to ground semantics of XML/XHTML documents? [RDFinXHTML-35]

From: <noah_mendelsohn@us.ibm.com>
Date: Wed, 20 Jun 2007 14:42:08 -0400
To: Pat Hayes <phayes@ihmc.us>
Cc: Dan Connolly <connolly@w3.org>, Ivan Herman <ivan@w3.org>, swick <swick@w3.org>, www-tag <www-tag@w3.org>
Message-ID: <OF08A3F888.AA56BA47-ON85257300.006637C3-85257300.006689F9@lotus.com>

I'm feeling dense here.  My understanding of the pertinent parts of the 
Web's Follow Yer Nose algorithm is:  get the Content-type from HTTP; find 
the specification for that content type;  the specification will tell you 
what semantics you can infer from the document.  So, to my naive reading, 
for RDFa to have full force in HTML, whichever content type you're using 
would have to say in it's specification:  if you see RDFa in the document, 
here's its meaning (presumably by delegating to the RDFa specifications.) 
How do DTDs help?

Noah Mendelsohn 
IBM Corporation
One Rogers Street
Cambridge, MA 02142

Pat Hayes <phayes@ihmc.us>
Sent by: www-tag-request@w3.org
06/20/2007 01:33 PM
        To:     Dan Connolly <connolly@w3.org>
        cc:     www-tag <www-tag@w3.org>, swick <swick@w3.org>, Ivan 
Herman <ivan@w3.org>, (bcc: Noah Mendelsohn/Cambridge/IBM)
        Subject:        Re: using DTDs to ground semantics of XML/XHTML 
documents?   [RDFinXHTML-35]

>The Semantic Web Deployment WG, working on RDFa, is
>considering an issue:
>How does one "Follow one's nose" from an HTML document to the RDFa spec?
>A recent proposal is, in short "through the DTD".

Seems to me that this makes sense only if the DTD is not just 
recommended for +RDFa, but *required* for it. Otherwise, even a 
bloodhound won't be able to get back from the document to the spec 
when the DTD is ignored or has gotten lost. Maybe I'm not following 
all the subtleties here.

What's wrong with the namespace way of doing it?


>I find that answer unappealing, though I don't have
>any particular argument against it. I noted my
>unease in response to the recent proposal.
>I wonder if anyone else has concerns about the use of DTDs, as opposed
>to namespaces, as a way to ground semantics of XML documents.
>This seems to be the approach used in XHTML Modularization.
>Dan Connolly, W3C http://www.w3.org/People/Connolly/

IHMC                             (850)434 8903 or (650)494 3973   home
40 South Alcaniz St.             (850)202 4416   office
Pensacola                                                (850)202 4440 fax
FL 32502                                                 (850)291 0667 
phayesAT-SIGNihmc.us       http://www.ihmc.us/users/phayes
Received on Wednesday, 20 June 2007 18:41:47 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:32:52 UTC