Re: Ongoing objection to RDFa Profiles format (as XHTML+RDFa)

Hi Nathan,

> Likewise I'll chime in with a big +1 for what you've both said, and having
> those profiles in (X)HTML+RDFa opens the door to a lot of functionality
> (indeed everything HTML, RDFa and related gives) so I feel it would be wise
> to open that door and promote wide spread RDFa adoption, after all if we
> can't mark up our own data in RDFa, how can we reasonably expect anybody
> else to?

I note that Richard also pursues this argument. :)

Are you and Richard now going to lobby for N3, SPARQL and RDF/XML to
also use RDF to define their prefix mappings?

Hopefully you'll mention to those communities that they can't
seriously expect anyone to use their technology if they're not even
prepared to mark-up their own prefix mappings with RDF?

Also, the assertion that we're marking up data ("marking up our own
data") assumes that the prefix mappings are marked up in RDF.

I've stressed *many* times that if we are going to use RDF to define
profiles then it's a no-brainer to use RDFa to serialise that RDF.

But my objection is to the use of RDF to define prefix mappings in the
first place!

It's an over-engineered solution that puts an unnecessary burden on
parser development, breaks the separation of concerns in the parser
architecture, and seems to meet only one requirement -- that when you
are carrying a hammer, everything looks like a nail.

> [snip...I agree with everything here. :) ]

> So Mark, I've got to ask, are you still against RDFa Profiles being in RDFa?

I'm not even thinking about profiles being in RDFa, because that's a
secondary question. But I remain very much against RDFa profiles being
in RDF!

> certainly seems as though we're all rather pro RDFa Profiles in RDFa at the
> minute, and that both of our concerns could be addressed with some notes..

Nope. :)

As I said in my blog post on 'Tokenising the Semantic Web' last year,
I can foresee a further development on profiles where we define an
algorithm that says how a vocabulary or ontology gets mapped to
tokens. With all due respect to everyone concerned my suspicion is
that this is far more along the lines of what people would like
profiles to be.

But that is not what we have at the moment -- we have a token-mapping
vocabulary which raises many problems without giving any benefit.



Mark Birbeck, webBackplane

webBackplane is a trading name of Backplane Ltd. (company number
05972288, registered office: 2nd Floor, 69/85 Tabernacle Street,
London, EC2A 4RR)

Received on Friday, 8 October 2010 11:43:46 UTC