W3C home > Mailing lists > Public > public-rif-wg@w3.org > January 2007

RE: Data models as constraints

From: Hassan Ait-Kaci <hak@ilog.com>
Date: Wed, 24 Jan 2007 06:15:41 -0800
Message-ID: <9FC9C6B2EA71ED4B826F55AC7C8B9AAB01F33573@mvmbx01.ilog.biz>
To: "Enrico Franconi" <franconi@inf.unibz.it>
Cc: "W3C RIF WG" <public-rif-wg@w3.org>
> From: Enrico Franconi [mailto:franconi@inf.unibz.it]
> Sent: Wed 1/24/2007 12:42 PM
> To: Hassan Ait-Kaci
> Cc: W3C RIF WG
> Subject: Re: Data models as constraints
> On 17 Jan 2007, at 20:49, Hassan At-Kaci wrote:
> > Enrico Franconi wrote:
> >
> >> I guess you have missed a few *crucial* references in your work:
> >
> > Thanks for kindly pointing these out to me. I am aware of most of them
> > as well of many others besides these. However, my goal is not a
> > *survey* on combining rules and ontologies, but to propose seeing data
> > models as constraints when used in rule-based schemes. Do these
> > "missed" references do so? (I do not think so - but I will check
> > again.)
> Sure they do so! DLs are about constraints, and the classical  
> approaches to integrate rules with DLs ara about that. So, I'd expect  
> that some comparison with the works done on this very same problem  
> should be considered.

I checked again, and I confirm that I was right as I wrote it: these
references (that you deem "crucial") do not use the CLP *scheme* for
their semantics. I agree (as I wrote it in my paper) that the DL-rule
langages using the ALCNR system of Bucheit et al., 1993 - i.e., CARIN,
AL-Log - use indeed a constraint-propagation method for proving DL
formulae. However, they do not make the link with CLP *at all*. This is
the extent of the relation to their work to mine, which I duly mention
in my paper (Sections 1.2, 3.3.3, and 4 - but you may have overlooked
reading these these sections after the references, perhaps?).

> > At any rate, you mean "crucial" in what sense?
> In the above sense.

Oh! So, I do not see anything crucial mising there at all... :-)

> > Do you imply that the paper is pointless without them? Does the fact
> > that I "missed" these references invalidate any of the contents? Or
> > make its message pointless? Or does "crucial" mean that these papers
> > are all of so momentous importance that no paper on rules for the SW
> > ought never to "leave home without it"? For me, a "crucial" miss is a
> > *serious* flaw. Is my paper seriously flawed?
> Cool down man, I was just saying that very similar goals have been  
> considered already in the literature, and that it may be useful for  
> your work to look at them. Maybe just to realise that your approach  
> is better anyway.

No sweat, buddy, I'm Kooool - with a Kapital K! :-) Thanks for the kind
advice anihoo: I get your drift and kinda dig your jive ... Thing is: it's
you, not I, that sounded all fluttered up that such "crucial" references
had been "missed"...

> > Anyway, thanks for reading through the thick of my slab so quickly.
> I don't see any smiley here :-)

Sorry: here they are: :-) ;-) :-O :-P :-D

> > PS/ This confirms that one always starts reading a paper with the most
> >     important section: the references! ;-)
> Is there anything wrong with that?

Hmmm ... Maybe not for you. Everyone has his/her own ethics. As for
me, I'd refrain from making peremptory comments based on so cursory
reading, or would mention that my reading - and hence my comments -
was thus shallow. I, for one, would surely think twice before making
such epithet-riddled statements as "you missed *crucial* references"!.

I understand that Latins tend to use stronger words than what they
intend to mean... Still!  ;-)

> cheers
> --e.



PS/ This exchange is no longer of interest to the public-rif list.
    So let's leave it at that, if you do not mind; or make it private.
Hassan At-Kaci
ILOG, Inc. - Product Division R&D
tel/fax: +1 (604) 930-5603 - email: hak @ ilog . com
Received on Wednesday, 24 January 2007 14:14:49 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:47:41 UTC