- From: Sandro Hawke <sandro@w3.org>
- Date: Thu, 01 Jul 2010 11:26:09 -0700
- To: nathan@webr3.org
- Cc: Jeremy Carroll <jeremy@topquadrant.com>, Yves Raimond <yves.raimond@gmail.com>, Pat Hayes <phayes@ihmc.us>, Toby Inkster <tai@g5n.co.uk>, David Booth <david@dbooth.org>, Dan Brickley <danbri@danbri.org>, Linked Data community <public-lod@w3.org>, Semantic Web <semantic-web@w3.org>, Michael Hausenblas <michael.hausenblas@deri.org>, Story Henry <henry.story@bblfish.net>, Kingsley Idehen <kidehen@openlinksw.com>, Ivan Mikhailov <imikhailov@openlinksw.com>
On Thu, 2010-07-01 at 17:39 +0100, Nathan wrote:
> Sandro Hawke wrote:
> > On Thu, 2010-07-01 at 17:10 +0100, Nathan wrote:
> >> In all honesty, if this doesn't happen, I personally will have no choice
> >> but to move to N3 for the bulk of things, and hope for other
> >> serializations of N3 to come along.
> >
> > RIF (which became a W3C Recommendation last week) is N3, mutated (in
> > some good ways and some bad ways, I suppose) by the community consensus
> > process. RIF is simultaneously the heir to N3 and a standard business
> > rules format.
> >
> > RIF's central syntax is XML-based, but there's room for a presentation
> > syntax that looks like N3. RIF includes triples which can have
> > literals as subject, of course. (In RIF, these triples are called
> > "frames". Well, sets of triples with a shared subject are called
> > frames, technically. But they are defined by the spec to be an
> > extension of RDF triples.)
>
> does it cover formulae in s and o position?
>
> i.e. can the following be expressed (without reification):
>
> { :thermostat :temp :high } log:implies { :heating :power "0" } .
It can express such rules. That's the main thing it does. It does not
consider rules to be triples, however. Making rules just be triples is
an interesting trick timbl did in the design of N3, but it causes a few
problems, and the RIF Working Group decided instead to make rules be
first-class syntactic objects instead. (The most obvious problem is:
where do you declare the variables? Another is: how do you attach
metadata to the rule? Many real-world rule systems have names and
other management information associated with each rule.)
As for putting formulas in the s and o position for non-rule
applications, I would argue that is, by definition, reification. There
is a RIF Working Draft [1] on doing that, but it's not part of the
current standard.
-- Sandro
[1] http://www.w3.org/TR/rif-in-rdf
Received on Friday, 2 July 2010 17:43:08 UTC