[whatwg] RDFa

[Ian Hickson]

On Tue, 9 Sep 2008, Philipp Serafin wrote:
>
> Please correct me if I'm wrong here, but it looks to me as if RDFa uses 
> namespaced identifiers nowhere outside attribute values right now. So 
> couldn't you just introduce a "rdf specific" namespacing system for 
> example like eRDF does[1]? This way, RDF parsers could still look up 
> metainformation about RDF properties while RDF-unaware agents could 
> parse the attribute values as simple text.

Oh I'm sure there are plenty of ways to handle the RDF problem space in 
text/html, assuming we establish that as being a problem worth solving in 
HTML5. I haven't had a chance to look closely at the feedback on RDF yet, 
there are a number of topic areas, in particular the forms, video, and 
offline features, that have a higher priority and have to be dealt with 
first.


On Tue, 9 Sep 2008, Ben Adida wrote:
> >
> > Both introducing a namespace prefix processing model and introducing 
> > DOM inconsistencies at the XML/HTML boundary intentionally are simply 
> > not an option in WHATWG specs at this point.
> 
> What is the evidence you have for making this decision, specifically 
> what is the evidence you have for conflating past issues with this one? 

A lot of the data on namespace issues is collected here:

   http://wiki.whatwg.org/wiki/Namespace_confusion

The edict against DOM inconsistency is primarily borne of feedback from 
browser vendors.


> Who gets to make this decision?

In the WHATWG the editor (me, for HTML5) makes the decisions, which can 
then be overriden by the core membership (as defined at the bottom of the 
charter) if they disagree with it:

   http://www.whatwg.org/charter


> What is the process by which such a decision is made?

The WHATWG process is described in the FAQ:

   http://wiki.whatwg.org/wiki/FAQ#The_WHATWG_Process


Note however that I do not expect the namespace issue to materially affect 
the RDFa feedback; I'm sure there are many ways of addressing the problem 
space of RDF that do not involve having to use namespace prefixes.

-- 
Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'