W3C home > Mailing lists > Public > www-rdf-interest@w3.org > November 2000

Re: Chainsaw?

From: Frank V. Castellucci <frankc@colconsulting.com>
Date: Wed, 08 Nov 2000 07:43:15 -0500
Message-ID: <3A094A63.D92A1E69@colconsulting.com>
To: Tom Van Eetvelde <tom.van_eetvelde@alcatel.be>
CC: RDF interest group <www-rdf-interest@w3.org>
Thats the interesting "problem" of semantics implied via the conceptual
graphs:

I could model

Car
 have colors
	Big Car
	Small Car

Company
	FordCompany
		FordCars
			FordEscort isA Small Car 
			colors = {Red, White, Blue,etc}

Or

Car
 have colors
	Big Car
	Small Car
		FordEscort isA FordCars

Company
	FordCompany
		FordCars

or probably a few other ways.

In the large, the conceptualization which enforce the semantics for
reasoning will have to whistle approximate tunes.

BTW: the colors enumeration specializes colors to restrict instances to
just "pick" from one of those, and of course the logic to enforce it.

-- 
Frank V. Castellucci
http://corelinux.sourceforge.net
OOA/OOD/C++ Standards and Guidelines for Linux
http://www.oods.org
Open Object Directory Services

Tom Van Eetvelde wrote:
> 
> Aha,
> 
> Does this mean the problem can be solved with statements and rules?
> 
> E.g.:
> 
> My statement:    Fordescort is defined by FORDCompany.
> My rule:             if x is a FordEscort, x has a color and it is red.
> 
> The rules define the context of FordEscort. The statements give extra info on the concept 'context
> of FordEscort'.
> 
> This goes into the direction of Prolog. Maybe one should ask oneself: is my modeling problem
> inherent to the problemdomain or am I using the wrong modelling tool? I guess if it is the latter,
> you can learn from the other modelling tool how to do things in your currently used modelling tool.
> 
> Greetings,
> 
> Tom.
> 
> Graham Klyne wrote:
> 
> > Jonathan,
> >
> > I for one don't know the answer to your question, but I'm trying to pursue
> > some ideas.  In considering and responding to Sergey's comments, I come to
> > think we're looking at different facets of the same problem:  I'm focused
> > on expressivity (with partial information), you're focused on inference, I
> > think.  In the end, I expect we'll find a solution that adequately captures
> > both, and hence answers your question.
> >
> > #g
> > --
> >
> > At 01:22 PM 11/2/00 -0500, Jonathan Borden wrote:
> > > > My interpretation of the second statement is that it says the *bag*
> > > > containing the statements that define a [FordEscort] is defined by
> > > > [FordMotorCompany].  My goal is to make that assertion about the (reified)
> > > > statements themselves.  (Consider, there may be another bag defined by
> > > > another party containing some of the same statements.)  The nearest thing
> > > > in the RDF spec is <Description 'aboutEach=...'>, but I find that lacks a
> > > > corresponding representation in the RDF abstract model.
> > >
> > >     And this is the problem. As much as it is good to keep the model as
> > >simple as is possible, the model needs to model the --isa--> chain properly.
> > >Part of this is the 'inheritance' of property values in a fashion similar to
> > >the way an object instance might 'inherit' const values defined as members
> > >of the class. Usually these const values can be static or class members. So
> > >the question is, how is this best modelled in RDF?
> > >
> > >Jonathan Borden
> > >The Open Healthcare Group
> > >http://www.openhealth.org
> >
> > ------------
> > Graham Klyne
> > (GK@ACM.ORG)
Received on Wednesday, 8 November 2000 07:39:21 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:51:46 GMT