Re: Meeting minutes, 2009-11-05

Toby Inkster wrote:
> On Thu, 2009-11-05 at 19:42 +0100, Ivan Herman wrote:
>>    ... There is MediaRSS - which is RDFa being used in a non-XHTML
>>    dialect.
> 
> DataRSS, surely? DataRSS is RDFa in Atom.
> 

I am sorry. Yes, I meant DataRSS. Thanks for catching this:-)

> Regarding an RSS 1.1 Core specification, it looks to me like there are
> currently three "branches" of RDFa syntax:
> 
> 1. XHTML+RDFa - the original, and a W3C Rec. DataRSS uses the same
> branch of RDFa - it's essentially just got different tag names.
> 
> 2. HTML+RDFa - small refinements on XHTML+RDFa necessary to get it to
> work on an HTML DOM.
> 
> 3. SVG 1.2 Tiny - this differs significantly from the other two in that
> it may also legitimately include RDF/XML chunks.
> 

Just for the records, AFAIK the differences are:

- SVG 1.2 accepts the genuine xml verbs like xml:base and xml:lang (and
the latter should be added to RDFa in general)
- SVG 1.2 restricts itself to the genuine RDFa attributes and does not
have special reference to @href or @src

which would probably characterize the generic RDFa for XML in general. I
would not consider these two as significant.

There is, of course, another item that you refer to below: SVG always
had the <metadata> element, and it was defined in such a way that this
could include <rdf:RDF> portions, ie, RDF/XML things. The way SVG 1.2 is
defined is such that the generated default graph is a merge of the RDFa
processing output and such RDF/XML statements. Which is indeed, as you
say below, a feature we might want to think about in general for XML in
general...

> RDF 1.1 Core 

You mean RDFa 1.1 and not RDF 1.1, right? :-)

>                needs to unite not only the first two, but also the third
> branch - XML-based varieties of RDFa which can legitimately include
> RDF/XML chunks. (It should specify whether triples from all of them form
> a single graph or should/may be separated into separate graphs. 

SVG 1.2 does not specify separate graphs

>                                                                  It might
> even be useful if:
> 
>  <x property="foo:bar" datatype="rdfa:reified">
>     <rdf:RDF>...</rdf:RDF>
>  </x>
> 
> resulted in something like the following N3:
> 
>  <> foo:bar { ... } .

Wow. That would open the floodgates vs. named graphs and reification...:-)

Ivan

> 
> Just some things to think about...
> 

-- 

Ivan Herman, W3C Semantic Web Activity Lead
Home: http://www.w3.org/People/Ivan/
mobile: +31-641044153
PGP Key: http://www.ivan-herman.net/pgpkey.html
FOAF: http://www.ivan-herman.net/foaf.rdf