W3C home > Mailing lists > Public > semantic-web@w3.org > August 2008

Re: rdf in xproc

From: Leigh Dodds <leigh@ldodds.com>
Date: Sat, 23 Aug 2008 10:47:30 +0100
Message-ID: <4393527c0808230247s51326c29gdadc9ee5fcafbf0b@mail.gmail.com>
To: "Holger Knublauch" <holger@knublauch.com>
Cc: "semantic-web at W3C" <semantic-web@w3c.org>, "Paul Tyson" <phtyson@sbcglobal.net>, ndw@nwalsh.com, "Jeremy Carroll" <jeremy@topquadrant.com>

> I think most people in the RDF community agree that operating directly on
> the RDF/XML syntax is not a perfect solution. For example there are many
> different renderings of the same RDF triples, so relying on XSLTs to do
> transformations is potentially very error-prone.

Absolutely, but its not impossible, especially when you're using constrained
versions of the RDF/XML syntax (which is which I meant by "subsets";
perhaps I should have made that clearer).

The point being that XProc, as an XML processing language, does let you
do some useful work with RDF/XML. The usual caveats about syntactic
variation will always apply.

> Having said this I very much agree with Paul's statement that having some
> standard pipeline facilities for RDF would be good thing.

Me too; But xproc isn't the basis for that. RDF would be better served by a
pipeline language whose basic unit is a set of triples rather an DOM
nodes. I've played with some simple approaches to this myself, although
nothing as complex as SPARQLMotion -- which looks very nice, and is
nicely named! -- more oriented towards simple compilation and construction
of collections of RDF data.

There's probably plenty of room for further exploration in this area.
But Paul's
original suggestion of, essentially, some xproc extensions would still be
useful.

Cheers,

L.
Received on Saturday, 23 August 2008 09:48:09 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 21:45:24 GMT