TEST: update

Jos

I've done a load of work on the doc.
I've added:
  - a few more tests (e.g. cardinality)
  - OWL Lite/OWL DL/OWL Full
  - conformance statements

The abstract syntax constraints are horrible - it's nearly impossible to
write OWL Lite or OWL DL by hand - take a look at the cardinality tests.
All I wanted to say was that
cardinality=1 =:= minCardinality=1 + maxCardinality=1

In OWL Lite (or OWL Dl for card=2)
(tests 001 002 003 004)
it takes 30 triples to say this.

Whereas in OWL Full it takes only 16 (test 005 006).

in short OWL Full is light and OWL Lite is heavy.

I moved to Jena2 (pre-alpha) because I needed some datatyping support.
You will need to download the jars in the lib dir in the editors draft.
The version on sealpc09 is currently down.
Also  near the beginning of CachedURLs.java I put a cheat to improve the
performance by first trying a local copy of the w3c cvs hierarchy.
It's got my directory path hardcoded at the moment, I assume you have one
too.

With the conformance statement I've currently commented out the old section
on running the tests. That's needs reinstating with some idea of which tests
should pass at which conformance level.

I am not sure if we need to talk about OWL Full conformance.
Peter says:
[[
The abstract syntax here is less general than the exchange syntax for OWL.
In particular, it does not permit the construction of self-referential
syntactic constructs. It is also intended for use in cases where classes,
properties, and individuals form disjoint collections.
]]
http://www.w3.org/TR/2002/WD-owl-semantics-20021108/syntax.html

An OWL Full system needs to be able to overcome these restrictions - even if
its just an editor.

I guess an OWL Full system also needs to be able to impose these
restrictions if OWL DL mode is in place.

I am going to work through the OWL Feature synopsis tomorrow starting with
the
sameXXXAs ...


Jeremy