W3C home > Mailing lists > Public > public-cwm-talk@w3.org > January to March 2010

Re: scope of _:existentials in N3Logic?

From: Yves Raimond <yves.raimond@gmail.com>
Date: Mon, 1 Mar 2010 09:57:53 +0000
Message-ID: <82593ac01003010157t560e0de8vd581b40cc3c474ee@mail.gmail.com>
To: Graham Klyne <GK-lists@ninebynine.org>
Cc: jos.deroo@agfa.com, connolly@w3.org, public-cwm-talk@w3.org, public-cwm-talk-request@w3.org

FWIW, I ran into some similar problems when playing with Swish and inference
> rules - handling bnodes got rather messy.  Eventually I figured (but didn't
> get round to implementing) that I could convert to some kind of unique
> generated id on loading the data, and use similar when introducing
> existentials in inferences, treat these like any other URI for inference
> purposes, than convert back to bnodes when saving the data.
When implementing the N3 reasoning part of Henry [1], I found that
skolemisation [2] was critical to make bnode handling relatively simple and
efficient - it is also fairly easy to implement. I believe Euler is also
using it.


[1] http://code.google.com/p/km-rdf/
[2] http://en.wikipedia.org/wiki/Skolem_normal_form
Received on Monday, 1 March 2010 09:58:26 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:01:06 UTC