W3C home > Mailing lists > Public > www-rdf-interest@w3.org > April 2003

RE: URI for language identifiers

From: <Patrick.Stickler@nokia.com>
Date: Tue, 1 Apr 2003 12:24:34 +0300
Message-ID: <A03E60B17132A84F9B4BB5EEDE57957B5FBB58@trebe006.europe.nokia.com>
To: <algermissen@acm.org>
Cc: <sandro@w3.org>, <dieter.koehler@philo.de>, <www-rdf-interest@w3.org>



> -----Original Message-----
> From: ext Jan Algermissen [mailto:algermissen@acm.org]
> Sent: 01 April, 2003 11:50
> To: Stickler Patrick (NMP/Tampere)
> Cc: sandro@w3.org; dieter.koehler@philo.de; www-rdf-interest@w3.org
> Subject: Re: URI for language identifiers
> 
> 
> Patrick,
> 
> Patrick.Stickler@nokia.com wrote:
> > 
> > > ... Given the RDF statement
> > >
> > > http://www.w3.org/Consortium/ http://foo/director "Tim 
> Berners Lee"
> > >
> > > doesn't 'http://www.w3.org/Consortium/' refer to the W3C itself?
> > 
> > You couldn't know that from the statement itself, in isolation.
> 
> Oh, but assuming that I understand the semantics of the foo namespace,
> I could - yes?

Not authoritatively. Any semantics conferred upon a subject by a predicate
is an implicit assertion by the author of the statement about the subject,
but not by the authority/owner of that actual subject. See below.

> Example: If the predicate is from Dublin Core, the subject is allways
> the webpage, never an abstract concept.

Well, actually, DC doesn't say that, neither explicitly by any domain
assertion nor in any of the prose descriptions/comments of the subject
property. So I don't see where you're getting that. But for the sake
of argument, let's just presume that there is something akin to

   dc:subject rdfs:domain ex:WebPage .

> I know, that is not RDF-ish thinking ;-)

Well, actually, it is. 

If a predicate has a domain defined for it, then it is quite OK to 
infer a type characteristic of any subject used with that predicate.
I.e., the use of the subject with that predicate is an implicit
assertion that the subject is of the particular type per the specified
domain.

*HOWEVER*, even if one may infer that implicit assertion based on
the domain of the predicate, that doesn't necessarily mean it is correct. 
The authority/owner of the subject may in fact not agree with such
an assertion and  there may very well be authoritative
information about the subject which conflicts with the assertions
inferred from its use with a given predicate.

No *use* of the subject with a given predicate counts as any authoritative
assertion about the nature of that subject.

If http://www.w3.org/Consortium in fact denotes a web page, no amount
of usage with the property http://foo/directory is going to change that
and make it denote an organization -- insofar as the authoritative
definition of the resource is concerned.

> > If the property http://foo/director had an rdfs:domain 
> defined such as
> > 
> >    http://foo/directory rdfs:range ex:Organization .
> > 
> > then one could be able to infer that 
> http://www.w3.org/Consortium denoted
> > the organization rather than the web page, since one could 
> then infer that
> > 
> >    http://www.w3.org/Consortium rdf:type ex:Organization .
> > 
> > But simply *using* the URI in a statement that presumes it 
> denotes the
> > organization (without actually being sure it does) is very 
> bad practice,
> > and does not in any way change the intended meaning of the URI.
> > 
> > If in fact, the owner of that URI asserts that
> > 
> >    http://www.w3.org/Consortium rdf:type ex:WebPage .
> > 
> > then using it as the subject of http://foo/director would 
> be an error
> > (presuming that web pages don't have directors, of course ;-)
> 
> Huh...does that mean that 'proper' use or RDF does not allow me to use
> addresses of existing web pages to refer to abstract 
> concepts? 

If those URIs denote web pages, then no, you certainly should not
use those URIs to denote any other resource, abstract or otherwise.

> That seems
> like a severe limitation to me? Of what use is an identifier 
> if I cannot
> use for example HTTP GET to 'see/read' what it means?

If a URI denotes an abstract concept, you may be able to GET a
representation of that resource. Why not.

But it is essential that each URI denote one and only one resource.

> Anyway, thanks Patrick, your reply is helping me a lot to understand
> RDF (I think).
> > 
> > > > (I made a proposal that RDF URI-Refs should sometimes be seen as
> > > > subject indicators [4], but it was not accepted.  After 
> trying to
> > > > implement my proposal I'm not so fond of it myself.)
> > >
> > > I think that proposal was quite good. IMHO the question if a URI
> > > identifies a 'page' or an abstract concept is NOT part of the
> > > semantics of that particular URI but depends on the linking
> > > context (the context in whitch the URI is used as a reference).
> > >
> > > But I never managed to get that idea across I think...
> > 
> > The idea that the context of usage affects the denotation of a URI
> > is one of the key points of incompatability between the TM 
> model and that
> > of RDF.
> > 
> > In RDF, a given URIref is expected to have a consistent 
> interpretation,
> > no matter what context it occurs in. 
> 
> But this means disconnecting the idea of URIs from HTTP, 
> right? URIs are
> then only unambigous names and not 'Web Addresses; any more. 
> I that true?

URIs are not "web addresses" per se.

For some URIs, one may use a protocol such as HTTP to obtain a representation
of the resource that is denoted by that URI, but one never gets the resource
itself.

In the case of digital resources, the representation returned by HTTP can be
a bit-equal copy of the digital resource, which gives the illusion that the
URI is acting as a repository or address of that digital resource, but there
is no requirement that a representation be bit-equal, and in fact content
negotiation exists to provide for multiple variant representations of a 
given resource.

So, for any given URI, it denotes a single resource. Some protocols, such
as HTTP, are able to work with particular URIs and provide a representation,
but whether or not any representation is available, or whether there is any
protocol at all that can do anything useful with that URI, the denotation
of the URI remains globally constant.

> 
> > No, that doesn't mean that that is
> > always case. 
> 
> > Ambiguity and bugs are part of real life on the SW. 
> 
> <rant>Are you sure that is a solid foundation to stand on?</rant>

I think that is the only reasonable position to hold. The Web and
the SW are pretty chaotic. The architecture should encourage good
practices, and should specify an ideal, but it is fairly unrealistic
to expect that the Web or SW will be perfect.

404 happens. So will disagreements about meaning.

Tools such as OWL will help us identify many of those disagreements,
and future machinery for scoping assertions will help us partition
knowledge into sub-webs of trust, but we will always be faced with
noise and garbage, and our SW agents will need to be designed to
deal with that gracefully.

That's reality.

> > But
> > consistent global meaning is the goal, and SW agents by 
> default should
> > be able to presume that whenever they encounter a given URIref, it
> > always means the same thing.
> > 
> > TMs, on the other hand, allow for variability in the 
> meaning of URIs,
> 
> No, in TM land, a URI allways is the address of 'the web page', a URI
> *never* addresses an abstract concept.

Well, that wasn't my understanding. But if that's true, then TMs and
RDF are even farther apart than I thought.

> Then in TMs URIs can be used as subject indicators, refering to
> arbitrary subjects. 

And how do you then make statements about the 'web page' versus
the subject? If you are using the same URI?

> A key concept is that when the URI of a 
> subject indicator
> is dereferenced and the retrieved information resource is 
> rendered for human 
> perception it should be clear what subject the URI indicates. 

But how do you differentiate between dereferencing the URI as
a subject indicator versus dereferencing the URI as a web page,
and is there any logical relationship between the web page
denoted by the URI versus the subject indicator denoted by the
same URI?

Having this ambiguity seems to make the core machinery alot more
complicated.

> > and are in that way at odds with RDF and IMO not fully nor safely
> > reconcilable with RDF graphs.
> > 
> > It's a pity, since there are alot of great ideas in the TM model. I
> > particularly envy TM scoping mechanisms and eagerly look forward to
> > the time that RDF has an analogous, and standardized, 
> scoping mechanism
> > for statements.
> 
> But couldn't you just make up a namespace that provides the 
> predicates to
> make scoping statements about statements???

Absolutely. Done that. Use it every day.

The key word is *standardized* ;-)

Cheers,

Patrick
Received on Tuesday, 1 April 2003 04:25:32 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:51:58 GMT