W3C home > Mailing lists > Public > www-rdf-logic@w3.org > June 2002

Re: Implementing statement grouping, contexts, quads and scopes

From: Patrick Stickler <patrick.stickler@nokia.com>
Date: Wed, 26 Jun 2002 11:00:18 +0300
To: Pat Hayes <phayes@ai.uwf.edu>
CC: RDF Logic <www-rdf-logic@w3.org>
Message-ID: <B93F4F42.1767E%patrick.stickler@nokia.com>


I think you are misunderstanding me. Please read the
following carefully, and remember, I may not always use
every word as you would ;-)

On 2002-06-25 19:45, "ext pat hayes" <phayes@ai.uwf.edu> wrote:

>> It can be simply a requirement on all OWL processors that they "pretend"
>> that unasserted RDF statements which contain OWL predicates are
>> asserted -- or part of an OWL preprocessor on an RDF graph which
>> maps unasserted OWL statements into asserted RDF statements. Whatever.
> This whole discussion isn't helped by getting ostentatiously vague
> and sloppy at the point where the problems arise. Look, if a 'higher'
> level language could be given a licence to 'pretend' that some piece
> of RDF means something different, of course there would be no
> layering problem.

I never (intentionally) said that upper layers would change the meaning
of lower layers.

The upper layer is simply augmenting the statements expressed at the
lower layer based on explicit rules (however those might be expressed
or executed).

The reification/description of the statement (the stating) remains
true and unchanged at *all* layers, but it is augmented with
an explicit assertion of that described statement by the higher
layer because it is recognized by the higher layer as having
special meaning that should be asserted at that higher layer.

The reification/description is simply a deferred assertion
mechanism, if you will. It doesn't change any meaning. It
just defers assertion of some triples until some higher layer

> That is exactly why we are suggesting a
> dark-triples mechanism, in fact.

I don't see this approach as incompatable with the idea
of dark-triples.

I think that perhaps you are understanding my proposal
as requiring that reification identifies a statement rather
than a stating, and if so you are not understanding me correctly.
What I am proposing should not require any substantial
change to how reification is presently defined, as describing
a stating.

A stating of course identifies a statement, and therefore
one can take any stating and extract and assert that
statement without any change to the meaning of the stating

It's like having a stating "John said the sky is green"
and a rule (e.g. at a higher level) that says "anything
that John says is true" which would assert all the
triples inherent in all statings by John. The fact that
the rule asserts those statements does not in any way
affect the meaning of the statings themselves.

Thus, at the RDF level, we'd have various statings using
OWL vocabulary. And at the higher OWL layer, we'd have
a rule that says, all statings using OWL vocabulary are
true (at the OWL level). Again, those upper layer assertions
do not change the truth of the statings themselves. The
statings are simply the mechanism by which assertion of upper
layer statements is defered until the graph is interpreted
by those upper layers.

So there are some things that are true at higher layers
that are not true at lower layers, but everything that is
true at a lower layer is also true at all higher layers.

If that doesn't capture the essense of dark triples, then I
must really be clueless.


Patrick Stickler              Phone: +358 50 483 9453
Senior Research Scientist     Fax:   +358 7180 35409
Nokia Research Center         Email: patrick.stickler@nokia.com
Received on Wednesday, 26 June 2002 03:55:49 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:52:42 GMT