Re: [office-metadata] [Fwd: Re: ODF and semantic web]

[Bruce D'Arcus <bdarcus@gmail.com>]

Svante Schubert wrote:

> Although we might even augment our set in later ODF specs, let's take a 
> closer look at adopting RDFa names.
> Especially as there is a new draft of RDFa available [1].
> 
> One ODF unfamiliar design is the absence of attribute namespaces. I 
> assume the reason for this design, was the expected greater acceptance 
> from the HTML community.
> Therefore my first question: would we use attributes without namespaces?

Yes, I had this question as well. Apparently, we're fine with 
namespacing them, though I confess this doesn't make much sense to me. 
It seems to me they (the W3C RDFa group) should define a namespace URI 
for the attributes, and create namespaced alternatives. We could then 
use those.

> Second aside of attribute names, there would be still some problem with 
> the attribute data types [2]. For instance @property uses only CURIEs, 
> where we use IRIs.

My understanding is the attributes in RDFa using either CURIEs or full 
URIs. Obviously, that would be necessary for us to be able to say we're 
a true subset.

Bruce