RE: Comments on RDF/A Syntax (Editor's Draft 27 October 2005)

Jeremy/Ben,

My first ever attempt at producing what became RDF/A actually did use
rdf:about [1]. In fact I tried to use *all* of the RDF/XML attributes, and
simply layer them onto XHTML. I was trying to make it so that a standard
RDF/XML parser could just start 'swallowing' RDF that it found in an XHTML
document. (An RDF/XML parser is supposed to skip over anything it doesn't
understand, so it's quite legitimate to place RDF/XML deep in a document.)

However, although I got close, it is actually not possible to do this (as
with a lot of things, it's the way predicates are done in RDF/XML that
ultimately catches you out). And in my view, unless the RDF attributes could
be used in their *proper* sense, they shouldn't be used at all. My
justification for this is not just because it might be confusing--although
that is certainly true--but also that just because I was not able to develop
a technique that incorporated *all* of the RDF/XML attributes into XHTML 2,
didn't mean that people couldn't still embed RDF/XML into XHTML documents in
the same way they could in any other documents. If we had co-opted the
RDF/XML attributes for our use but slightly changed the rules, it would
cause mayhem!

The next iteration on the path to RDF/A (called RDF/XHTML [2]) therefore
involved adding attributes that were specific to XHTML 2.

Regards,

Mark

[1] http://www.formsplayer.com/notes/xhtml-meta-data-02.html
[2] http://www.formsplayer.com/notes/xhtml-meta-data-03.html


Mark Birbeck
CEO
x-port.net Ltd.

e: Mark.Birbeck@x-port.net
t: +44 (0) 20 7689 9232
w: http://www.formsPlayer.com/

Download our XForms processor from
http://www.formsPlayer.com/  

> -----Original Message-----
> From: public-rdf-in-xhtml-tf-request@w3.org 
> [mailto:public-rdf-in-xhtml-tf-request@w3.org] On Behalf Of Ben Adida
> Sent: 27 November 2005 22:42
> To: Jeremy Wong ???
> Cc: public-rdf-in-xhtml-tf@w3.org
> Subject: Re: Comments on RDF/A Syntax (Editor's Draft 27 October 2005)
> 
> 
> 
> Jeremy,
> 
> Thanks for your comments. A quick response below.
> 
> > 1. "rdf:about" instead of "about"
> > "rdf:about" is pretty well-known. To learn a new attribute, 
> > "rdf:about" and "about" are no difference. Just a name to 
> remember...
> > e.g., <p rdf:about=""><a rel="dc:creator"  
> > href="mailto:bill.gates@microsoft.com">Bill Gates</a> blah 
> blah</p> => 
> > { <> dc:creator <mailto:bill.gates@microsoft> . }
> 
> Having rdf:about but xhtml2:rel, xhtml2:property, and 
> xhtml2:href seems inconsistent. Having them all scoped as 
> rdf: is not possible without adding to that namespace, which 
> seems like overkill.
> 
> Note also that rdf:about is completely unknown to the HTML 
> community, so I suspect this would cause more confusion than 
> consistency.
> 
> > 2. use the "type" attribute for typed literal The "type" 
> attribute is 
> > from the anchor element.
> > e.g., <p rdf:about=""><a rel="dc:date"  
> > type="xsd:dateTime">2005-11-25T00:46:00+0800</a> blah 
> blah</p> => { <> 
> > dc:date "2005-11-25T00:46:00+0800"^^xsd:dateTime . }
> 
> So you're suggesting "type" instead of "datatype". I'm not 
> expert enough on this issue to say which is better. Any 
> particular reason for your suggestion?
> 
> > 3. use of Production "propertyAttr" [1] I recommend to use 
> > "propertyAttr" for the case of "nodeElement" [2] only, never 
> > "emptyPropertyElt" [3] e.g. <p rdf:about="" 
> > dc:subject="SemanticWeb">blah blah blah</p> => { <> dc:subject 
> > "SemanticWeb" . }
> 
> This is an issue which we've currently put off. In the 
> current draft, a property is never declared as an attribute 
> except in special HTML- specific cases (like class and role). 
> We may take this up again, though right now we're leaning 
> towards simpler, fewer methods of saying the same thing.
> 
> -Ben
> 
>

Received on Sunday, 27 November 2005 23:09:50 UTC