Re: [RDFa] ISSUE-28: following your nose to the RDFa specification

On Mon, 2007-06-18 at 17:00 -0700, Ben Adida wrote:
> 
> Issue #28:
> http://www.w3.org/2006/07/SWD/track/issues/28
> 
> DanC asked us: "how does one follow one's nose to the RDFa specification
> from an HTML+RDFa document?"
> 
> I propose that we respond to this question by pointing to the DTD
> declaration we now recommend for XHTML1.1+RDFa documents.
> 
> I also propose that we specify an official W3C profile for XHTML1.1+RDFa
> which would include a GRDDL transformation for XHTML1.1+RDFa. We should
> encourage publishers to use this profile when it's possible, though we
> should not to developers of RDFa consumers that this profile may not
> always be present, since RDFa is built for copy-and-pasted, widgets, etc...
> 
> Thoughts? Comments? Remember, send a note even if you simply agree!

Hmm... surely the copy-and-paste caveat applies to the DTD as well, no?
Is the DTD optional? The caveat seems to apply to the MIME
type too ("In the case of XHTML1.1+RDFa, application/xhtml+xml.")
If somebody can't change the top of the document,
I find it hard to believe that they could change the MIME type.

I'd sure like to see clarity on requirements such as "RDFa markup
must work when pasted in the middle of an HTML-as-she-are-spoke
document". I think it's best to be clear that this is a goal
of RDFa, but we don't expect to fully meet the requirement
until in-progress HTML specs mature.

I support specifying a profile, without reservation.

I don't support using DTDs as part of the follow-your-nose
trail; they're not visible from XPath, I don't think;
are they visible from javascript? SAX has a relevant event,
but I have some vague feeling SAX parsers don't have to fire it.
If others are satisfied, I suppose I can live with it,
but it doesn't appeal to me at all.

Way back when, I asked in comp.text.sgml whether DTDs were intended
to convey semantics this way, and I was convinced by Erik Naggum
that they are not.

Ah yes... relevant parts of the discussion are still available...

"There are many ways to standardize semantics and to allow people 
to share information in useful ways.  Standardizing DTD's is not among
them."
 -- Eric Naggum, Nov 9 1993
http://groups.google.com/group/comp.text.sgml/browse_thread/thread/16158877bf81979d/dcee2bb9e79da4ad?lnk=st&q=comp.text.sgml+naggum+dtd+semantics&rnum=4#dcee2bb9e79da4ad

-- 
Dan Connolly, W3C http://www.w3.org/People/Connolly/

Received on Tuesday, 19 June 2007 11:33:07 UTC