> -----Original Message-----
> From: ext Roland Schwaenzl
> [mailto:[log in to unmask]]
> Sent: 17 January, 2003 11:11
> To: [log in to unmask]
> Subject: Re: rdfs:isDefinedBy revisited
>
>
> I don't see what this has to do with the semantic web
> > > > > (apart from it
> > > > > being part of the web).
> > > >
> > > > Because most folks think that a URI denotes what they get
> > > > when they dereference it, and that is not necessarily what
> > > > it actually denotes to a reasoning engine. Yet if they
> > > > start making statements about what they got, using the
> > > > name of "where" they got it, then they introduce noise
> > > > into the SW.
> > > >
> > > > If I have a URI that denotes the city of Paris, and someone
> > > > dereferences it and gets a representation that is a photo
> > > > of Paris, and they say "Paris is out of focus" when they
> > > > really meant to say, "The photographic representation of Paris
> > > > I got is out of focus" then the SW becomes a repository for
> > > > garbage, not knowledge.
> > >
> > > I don't quite get what you want to say.
> > >
> > > Suppose i retrieve a picture from a URL1, which according to my
> > > previous knowledge shows a cat and you assert in some
> > > semantic web URL1 --rdf:type--> "dog"
> > > Can you tell what conclusion i should make in your semantic web?
> >
> > That representations are not a reliable basis for making
> > statements about resources (at present).
>
>
> ...a statement which brings me back to Jon's interjection:
>
> I don't see in which respect this is a specifics of the semantic web
> (apart from it being a part of the web)....
>
> rs
It is not specific to the semantic web, in isolation, but
a component of the recent thread included the interface
between the SW and the Web, as to the relationship between
what a URI denotes and what one GETs when one dereferences
that URI.
Thus, if an XML namespace URI denotes the namespace, and
when I dereference it I get an XML Schema that defines
a information model (which just happens to use terms in
that namespace) or I get a RDDL instance that describes
arbitrary resources having some relation to terms in
that namespace -- are those valid representations of the
namespace itself, and does that not encourage the misinterpretation
of the namespace URI also denoting the document model, or
the exact schema or the RDDL instance, etc.
Insofar as the SW is concerned, a URI need never be
dereferenced. But SW *agents* will regularly dereference
URIs denoting certain classes of resources in order to
utilize representations of those resources, and the
success of the SW within the context of the Web depends on
being clear about (a) what those URIs actually denote, (b) that
what is returned is a representation of what is actually denoted,
(c) that a URI denotes only one thing, and (d) that the "default"
representation is "reasonable" in terms of the resource
denoted (i.e. I don't get an MPEG video of someone talking
about a document model when I dereference a URI denoting
a specific XML Schema instance but rather, some form
of semantically complete representation of the schema
itself, etc.).
Anything less, and all these discussions about XML namespaces,
DCMI namespaces, vocabularies, terms, resources, representations,
etc. remain simply academic and cannot be put in to practical
consistent use. I.e. these need to be addressed by a standard
by which SW agents can be constructed.
But as I've said before, this particular discussion has
probably strayed outside the normal scope of this particular
discussion list...
Patrick
|