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

RE: DAML Level of Effort for FY03-FY05

From: Didier VILLEVALOIS <dvillevalois@techmetrix.net>
Date: Mon, 22 Apr 2002 18:25:47 +0200
Message-ID: <6CED7AD2680ED611BB8E0090276CE92401DAA425@mail-paris.sqli.com>
To: Jeff Heflin <heflin@cse.lehigh.edu>, Drew McDermott <drew.mcdermott@yale.edu>
Cc: drager@bbn.com, www-rdf-logic@w3.org
Hello,

Shouldn't RDF support something like daml:imports in its core ??
daml:imports would be a subPropertyOf 'rdf:imports'.
As rdf enables to speak about the same things in different places, it needs
a such mechanism.

This would enable everybody to make rdf documents that are perspectives on
the semantic web.
I think this should not be as high-level as being part of DAML, because
everybody's semweb engine may not have it.

Didier.

> -----Message d'origine-----
> De: Jeff Heflin [mailto:heflin@cse.lehigh.edu]
> Date: lundi 22 avril 2002 18:11
> : Drew McDermott
> Cc: drager@bbn.com; www-rdf-logic@w3.org
> Objet: Re: DAML Level of Effort for FY03-FY05
> 
> 
> Drew,
> 
> Since I was the initial proponent of daml:imports on the Joint
> Committee, let me address this issue. You are absolutely correct that
> the imports statement must be used. Simply refering to a 
> namespace does
> not include any ontology definitions. You must make the imports
> statement explicit. Period.
> The correct way to do this for instance data is:
> 
> <rdf:Description about="">
>   <daml:imports
> rdf:resource="http://www.daml.org/experiment/ontology/beta/ele
> ments-ont#"/>
> </rdf:Description>
> 
> The problem with using RDF namespaces to decide which schemas are
> relevant is that multiple files may contain different 
> definitions about
> the same URI. See the attached GIF for example. The URI for 
> Dolphin has
> additional definitions in two schemas, good-schema and 
> bad-schema. These
> definitions are simply rdfs:subClassOf statements that happen to have
> orig-schema#Dolphin as their subject. The problem with simply using
> namespaces is I can't say that my-doc agrees with the definitions of
> Dolphin found in good-schema but not those found in 
> bad-schema. This is
> why daml:imports was an essential component of the language. Note that
> this issue is also addressed in the Web Ontology Requirements document
> [1].
> 
> Now this does mean that there will be some apparent redundancy between
> the namespace declarations and the imports declarations in any DAML
> documents. This is unfortunate, but is the result of XML namespaces
> being defined solely to solve the namespace problem for 
> languages before
> the W3C was concerned with ontologies. The ideal solution would be to
> have the imports introduce namespaces for each ontology (as 
> is done with
> most pre-Web ontology languages), but alas this solution would go
> against the grain of pre-existing web standards.
> 
> Theoretically we could say that the use of a namespace 
> implies that you
> import that ontology, but this does have the issue (raised 
> later in this
> thread) of distinguishing ontologies from other XML 
> namespaces. Also, it
> would not allow someone to provide completely alternate 
> definitions for
> a common set of terms (whether this is a good thing or a bad 
> thing, I'm
> not sure). Regardless, there is nothing in the DAML+OIL specs 
> that says
> that such default import statements exist. Although some tools may
> decide to implement this default, it is nonstandard, and is akin to to
> having HTML tools that are very flexible about guessing what you mean
> when you provide malformed HTML.
> 
> Jeff Heflin
> 
> [1]
> http://www.w3.org/TR/2002/WD-webont-req-20020307/#goal-shared-
ontologies

Drew McDermott wrote:
> 
> Hello, Dave --
> 
> In your capacity as keeper of the DARPA/SONAT ontologies, I have a
> question about the "instance data."  I am cc-ing www-rdf-logic because
> there seems to be a general issue here about how DAML datasets work.
> 
> These ontologies are found at
> 
> http://www.daml.org/experiment/ontology/
> 
> and I don't think they're sensitive (or they would be https: instead
> of http:).
> 
> The file af-a.daml is RDF, rather than DAML, and appears to consist of
> "instance data," not ontology.
> 
> Here is the beginning of that file:
> 
> <rdf:RDF
>   xmlns:rdf='http://www.w3.org/1999/02/22-rdf-syntax-ns#'
>   xmlns:rdfs='http://www.w3.org/2000/01/rdf-schema#'
>   xmlns:NS0='http://www.daml.org/experiment/ontology/beta/elements-ont#'
>  >
> 
>   <rdf:Description
rdf:about='http://www.daml.org/2001/12/factbook/af.daml#A1202'>
>     <NS0:country
rdf:resource='http://www.daml.org/2001/09/countries/fips#AF'/>
>     <rdf:type
rdf:resource='http://www.daml.org/2001/12/factbook/factbook-ont#Port'/>
>     <rdfs:label>Kheyrabad</rdfs:label>
>   </rdf:Description>
> 
> My problem is this:
> 
> Wouldn't it be much more informative if this line
> 
>   xmlns:NS0='http://www.daml.org/experiment/ontology/beta/elements-ont#'
> 
> were supplemented by something like this:
> 
>   <daml:imports
rdf:resource="http://www.daml.org/experiment/ontology/beta/elements-ont#"/>
> 
> But it's not clear where this would go.
> 
> Is it supposed to be the case that just mentioning an ontology's
> namespace means that the file uses that ontology?  I find that rather
> disconcerting.  For the time being, we'll make that assumption, but
> this seems like a bug (or gap) in the DAML spec.  There ought to be an
> official way to say "This RDF file is based on ontologies X, Y, and
> Z."
> 
>                                              -- Drew McDermott
Received on Monday, 22 April 2002 12:26:24 GMT

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