Re: resources and URIs

pat hayes wrote:

> On this picture, the information (which Dan, in your introductory 
> example, reads on his screen, and which is in some sense all about the 
> weather in Oaxaca) is a representation of the (current state of) some 
> entity *in the WWW itself*: a resource in the global information 
> network: the state of some computer system, or maybe some abstraction of 
> a computer system.

The first half is correct; if it has a URI it's part of the Web, if you
can get representations it's a more useful part.  What the resource *is*
in this example is not specified.

If it pleases you to assert that the URI identifies "Oaxaca's weather
forecast according to the example.com online weather service" then this
is a useful representation.  For the Web to behave well, we should be
aware of what the owner of example.com asserts that this resource is,
although this doesn't really constrain the way we interpret any
representations we get back.

> However, it is also clear that neither the weather in Oaxala, nor Oaxala 
> itself, are entities of this kind:  weather and cities in Mexico are not 
> the kind of entities which can be thought of as 'objects on the 
> networked information system'. 

On the contrary.  If the proprietors of example.com assert that
http://weather.example.com/oaxaca is intended to represent the weather
in Oaxaca, as an abstraction, for use in the development of a formal
inferencing system about global weather, and its primary practical use
is to retrieve slots from a KR repository, well, hey, that's OK too.  In
this case it would be surprising if the proprietor provided much in the
way of representations, and if they did, they would likely be quite
different from those provided if the operator asserted it was an online
weather service.

The Web Architecture doesn't provide a way to talk about what resources 
are or what they mean.  For the moment, we're stuck with human language 
for that.  The Web provides a way to name them and a way to move 
representations around.  That's all the Web of today needs.

The semantic web, one would hope, will provide a way to talk about what 
resources mean and the relationships between them.  But it had better be 
able to work usefully with the Web as it is.

> Now, this could be just a matter of philosophical opinion, were it not 
> for the fact that semantic web languages like RDF and OWL have been 
> given *formal* semantic theories which have direct architectural 
> consequences for Web agents, and which depend crucially on notions like 
> the term 'about' I have used rather loosely above.  RDF uses URI 
> references as *names* to *refer* to entities. So if a web page such as 
> http://chandra.harvard.edu/photo/2003/ngc1068/index.html were to include 
> RDF markup, one might expect to find things like this in it:

This is all very well and good, but it is neither necessary nor 
appropriate that we pretend that the current architecture of the web 
comprises any notion of "about-ness".  It has URIs, representations, and 
protocols to move them around.  You may not like this, but you have to 
deal with it, because that's just the way it is.

> As a result, several of the assertions you make in this document are not 
> correct. For example
> 
> 2.8.2
> "merging Semantic Web technologies, including "DAML+OIL" [ DAMLOIL ] and 
> "Web Ontology Language (OWL)" [ OWL10 ], define RDF properties such as 
> equivalentTo and FunctionalProperty to state -- or at least claim -- 
> formally that two URIs identify the same resource. "
> 
> is incorrect. These assertions claim that two URI references *denote* 
> the same entity in all interpretations. That is not the same notion as 
> 'identify'.

I think we'd be happy to accept corrections if our statements about what 
SW technologies do are incorrect.  To be useful, it will have to be more 
elaborated than that above, because I suspect that most, like me, will 
not have any idea what the differences are between identification and 
denotation.

> 
> In fact, there is no such notion as 'identify' in RDF/RDFS/OWL 
> semantics; and the first principle in section 2 ("All important 
> resources SHOULD be identified by a URI ") is meaningless when taken 
> literally in the context of semantic web languages

So what?  It is a correct and important assertion (and one that needs to 
be made) that you're not part of the Web until you have a URI, and that 
being a part of the Web is a good thing.  If there is a problem with 
taking it literally in the SW context, then don't take it literally in 
the SW context.  This worries me, because the statement is correct and 
useful.

>, as URIs there 
> typically cannot be said to identify anything: they act as names whose 
> possible referents are constrained by the assertions made using them, 
> but they are not 'linked' to anything, not 'bound' to anything, and are 
> not obliged to 'identify' anything; and the universes of discourse may 
> contain entities which cannot possibly be all identified or even 
> referred to by URIs, since there are too many of them, or it is 
> physically impossible to identify them with enough precision, or simply 
> because it is impractical to do so.

I don't understand the above.  URIs in the Web architecture are what 
they are, and what they are is effectively defined by a huge universe of 
deployed technology; they are character strings that can be used to look 
things up in databases and to retrieve representations.  This is the 
basis of the Web architecture.  If you want to use them as a basis for 
building the SW, that's fine, but please don't try to stop us from 
writing down an accurate description of reality as it empirically is.

> We need to get clear on this issue, or else we will continue to be mired 
> in confusion.

We are *entirely* clear what URIs are in the context of the Web-that-is, 
and we have the software to prove it.  We are trying to write down that 
shared understanding.

> 
> Let me suggest that it would be worth distinguishing between what a 
> representation is *about*, and what resource *produced* it. 

This sounds like an excellent job for the Semantic Web.  The current Web 
gets along fine without such a distinction.

> The document 
> currently says that URIs are used to retrieve representations 'of' a 
> resource.  It is easy to read this as saying that the representation is 
> 'about' the resource: that it 'refers to' or 'describes' the resource; 
> but this is evidently incompatible with the notion of a resource as 
> something that must be 'part of' an informational network.

You can read anything you want into it, but it doesn't actually say 
anything more than what you describe in the first sentence.  Nor should it.

--
Cheers, Tim Bray
         (ongoing fragmented essay: http://www.tbray.org/ongoing/)

Received on Tuesday, 15 July 2003 20:54:43 UTC