Introduction

Hello,

I'm the Working Group's newest invited expert.

I work part time for a charity in the UK and take on occasional freelance
work, though my interests in the Semantic Web are mostly a hobby.

A couple of years ago I started development of a Perl parser for
microformats and other semantic HTML formats (e.g. eRDF, DC-HTML, etc).
Development of the parser, Swignition, is now mostly stopped, though I'm
cannibalising it to split out some of the more interesting components.
RDFa was one of the formats I implemented, and it quickly became my
favourite.

The RDFa parser was the first thing to be split out from Swignition, and
is available on CPAN under the name RDF::RDFa::Parser. It already
implements (as options that can be selectively enabled) many of the
features that are on the agenda for this working group, including support
for the 'lang' attribute, support for non-XHTML host languages, support
for full URIs in CURIE attributes, and various ways of handling
prefix-less CURIEs.

Another project I'm working on (on and off) is my RDF-based CMS, demiblog.
It keeps all its data in a triple store and serves it up using
machine-generated RDFa.

My work items of interest for the RDFa WG are:

1. Keeping RDFa 1.1 compatible with RDFa 1.0. I'm defining "compatible"
thusly: any document authored as RDFa 1.0 should produce the same graph
when parsed with an RDFa 1.1 parser; and any document authored as RDFa 1.1
should produce the same graph, or a proper subgraph when parsed with a
legacy RDFa 1.0 parser.

2. Try to ensure that RDFa external vocabs, or whatever they end up being
called, don't end up making RDFa documents more fragile; don't break too
badly when the vocab document disappears; don't introduce centralised
points of failure.

3. Trying to reduce divergence in how RDFa is adopted by different host
languages. RDFa Core 1.1 should specify which features of RDFa host
languages may alter, and which they should not.

4. It would be nice (if people think it's within the group's scope) to
look at publishing something like Named Graphs in RDFa
<http://buzzword.org.uk/2009/rdfa4/spec> as a Working Group Note.

-Toby

Received on Wednesday, 3 March 2010 10:09:30 UTC