W3C home > Mailing lists > Public > public-owl-dev@w3.org > July to September 2011

Re: A solution to integrate CWA into OWA.

From: Pat Hayes <phayes@ihmc.us>
Date: Fri, 19 Aug 2011 16:54:20 -0500
Cc: <public-owl-dev@w3.org>
Message-Id: <56A4189A-78C7-4A89-B6E2-FF9683596AD2@ihmc.us>
To: duanyucong <duanyucong@hotmail.com>

On Aug 19, 2011, at 4:22 PM, duanyucong wrote:

> Dear Pat,
> If you kindly allow me to discuss a little more on the direction which you might mean by "......far-reaching metaphysical dimensions that you seem to think it has. ..."
> I would like to view from perspectives of semantic level, notation level and conceptual level.

Well, part of our problem is that we do not understand what you mean by these 'levels'. The logical picture behind OWL and similar languages is that the names in the logic refer to things in the world, and that the assertions in the logic express facts, and ground facts have the content that two or more things stand in a certain relationship to one another. This is what 'semantics' means here: it means, this mapping from the expressions to a world which the expressions describe and the names refer to.

> (semantic refers to the meaning of an expression in individuals's minds; 

Already, you are talking about something else, it seems. There is nothing 'psychological' in the semantic picture I outlined above. Minds do not enter into the picture at all.

> notations are the symbols used for expression;

OK, we agree on that :-)

> concepts are elements of expressions which composes notations and semantics.

I confess to simply not being able to understand what you mean here. But in any case, talk of 'concepts' is rather beside the point of standard logical semantics. 

> )
> Then negation mean different from these perspective. Part of what i would like to say outside OWL is that: at semantic level, the negation/"~" in OWA  is not completely the same as the negation/"~"  of  CWA.

Hmm. Well, certainly, one can *distinguish* two different notions of negation, and indeed this has been suggested and is actually done in several formalisms. One is 'classical' negation, usually written as the word NOT, or indicated by a tilde sign ~, which simply means that the truth-value is inverted: NOT P is true just when P is false and false just when P is true. Then 'failure negation', sometimes written as the word NAF, has a different meaning. NAF P means that some kind of systematic attempt to prove P has failed to establish its truth. And then the closed world assumption for P can be stated explicitly as the implication 
(NAF P) implies (NOT P). 
Of course, this sketchy account has to be made more concrete by filling out the missing details. (What counts as a failure? What process is involved? And so on.) But yes, we can certainly make this distinction between two different notions of negation; since they are, indeed, different.

Have I followed you so far?

BTW, there are some articles in Wikipedia which together give a good survey of the classical results in this area, with pointers to the relevant readings from the mid-1980s. I would suggest doing some reading before venturing further, in order to get better acquainted with the state of the art.

Pat Hayes

> Sincerely,
> Yucong
> Subject: Re: A solution to integrate CWA into OWA.
> From: bparsia@cs.man. ac.uk
> Date: Fri, 19 Aug 2011 12:26:14 +0100
> CC: public-owl-dev@w3.org
> To: duanyucong@hotmail.com
> I'm going to be blunt: AFAICT, most of what you've written is gibberish. I say this not to insult, though that's hardly avoidable, but to make clear how off base what you've written is. It may have its roots in some idiosyncratic terminological choices you make...I don't know.
> On 18 Aug 2011, at 22:27, duanyucong wrote:
> Dear all,
> I think that there is usually a misunderstanding on the meaning of CWA vs. OWA:
> These are really well understood notions with reasonably precise technical meanings. I know people often *are* confused, but I don't think there's a systematic confusion in the literature.
> This is a warning sign.
> It is  because of that a discussion might be extended based on implicit understanding of the semantic possibilities of these two concepts.
> for the cases like "...Therefore facts not stored in the database and not derivable from the existing data 
> are considered false in the CWA and unknown or possible in the OWA.... " 
> ---http://www.dsc.ufcg.edu.br/~ulrich/Artigos/MITO SBBD97.pdf 
> This is a perfectly reasonable, if not ideal, account.
> My argumentation:&n bsp;
> 1. in OWA, negation is not "considerated" at all.
> Regardless of the scare quotes, this is trivially refutable.
> 	C => ~D.
> 	~D => E.
> entails
> 	C => E.
> Negation is fully considered.
> 	C & ~C => owl:Nothing.
> Negation is fully (and actively) considered. I.e., unlike the first one, you cannot treat the negation as part of the atomic concept and preserve reasoning.
> 	a: C&~C.
> is inconsistent.
> Negation is fully considered.
> This is obviously standard first order negation.
> Or in another word, negation is not cognitively available in the mind, and subsequently not available in semantic expressions produced/organized in the mind. 
> This is really quite meaningless even on congitive science terms, but is obviously unrelated to the technical, mathematical aspects of OWA/CWA.
> Cheers,
> Bijan.

IHMC                                     (850)434 8903 or (650)494 3973   
40 South Alcaniz St.           (850)202 4416   office
Pensacola                            (850)202 4440   fax
FL 32502                              (850)291 0667   mobile
phayesAT-SIGNihmc.us       http://www.ihmc.us/users/phayes
Received on Friday, 19 August 2011 21:54:51 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:58:20 UTC