Hacking RDF with Perl hackathon cancelled

[Toby Inkster]

On Sun, 18 Apr 2010 21:48:39 +0200
Dan Brickley <danbri at danbri.org> wrote:

> What are the main opensource pure Perl tools available? 

RDF::Trine and RDF::Query tend to be the framework we're gravitating
towards. For XML parsing, RDF::Trine has an XML::LibXML dependency, but
other than that it's fairly pure Perl.

> How compliant
> are they? How much of SPARQL (in memory, in SQL, in other things;
> SPARQL protocol; SPARQL 1.1?) is covered?

RDF::Query has very good SPARQL 1.0 support, and some SPARQL 1.1. It
can operate on in-memory and database-backed models.

Additionally, RDF::Query::Client provides an interface for
SPARQL-protocol endpoints, using an almost identical API, so you can
switch between a local model and a remote endpoint very easily.

> RDFa? GRDDL? OWL? RIF? N3? POWDER?

My RDFa parser passes the RDFa 1.0 test suite, and I'm working on
experimental RDFa 1.1 support. If my parser's installed, RDF::Trine
will pick it up and use it.

GRDDL I have on my TODO list.

One of the things we were hoping to work on at the workshop was a
rule-based forward-chaining reasoner, with a ruleset for RDFS. Greg
already has one in his git repo, but it needs some cleaning up for
release. I imagine support for a large subset of OWL would be pretty
easy to add too, and vocab-specific reasoning (e.g. for FOAF, generate
mbox_sha1sum automatically from mbox properties).

I don't think anyone has anything planned regarding N3, RIF or POWDER
yet.

> Are the tools happily packaged in CPAN? (and Debian etc, if that's
> needed?). Are they well tested, actively maintained?

Everything I've mentioned above is on CPAN except the reasoner.

-- 
Toby A Inkster
<mailto:mail at tobyinkster.co.uk>
<http://tobyinkster.co.uk>