RE: AWWSW telecon, Tues Feb 3

> From: Jonathan Rees [mailto:jar@creativecommons.org] 
> 
> How would this be useful?

This group has been struggling for over a year to understand and explain what an "information resource" is and what it really means when a 200 response is returned upon dereferencing a URI.  These questions boil down to fundamental architectural questions of how identity is determined and what it means for a URI to "denote" a resource.  The kind of ambiguity described below is specifically the kind that is causing so much consternation: it cuts to the heart of why 

> 
> Ambiguity in our context can be treated with RDF and OWL semantics:  
> you can have multiple models of the same set of axioms, and 
> the choice  
> of model is up to whoever is trying to interpret the axioms. 

Sure it can be, but the result is that one doesn't gain the insight that could otherwise be gained.  Punting the problem off to a little cloud labeled "interpretation" doesn't help.  The AWWW says
http://www.w3.org/TR/webarch/#id-resources
"By design a URI identifies one resource".  But the fact is, a URI does *not* always identify only one thing.   Ambiguity is inescapable, so we don't get ourselves very far by trying to punt on what it means and how we can deal with.

David Booth
  

> Meaning  
> is expressed formally to the extent you have the cleverness and  
> endurance to formulate it that way, and using rdfs:comment (or  
> specification documents) otherwise; the former gives a way to 
> exclude  
> interpretations, while the latter might influence someone to choose  
> between the ones that remain. 
> I would prefer we stick with 
> the problem  
> of figuring out what axioms should be put into the inference pot,  
> based on the HTTP (REST, etc.) protocol, rather than get into the  
> problem of modeling RDF or OWL semantics in RDF or OWL, or inventing  
> protocols for communicating, attributing, authorizing etc. intended  
> denotation, which might be interesting but as far as I know 
> are not on  
> the table here.
> 
> On Jan 30, 2009, at 2:47 PM, Booth, David (HP Software - 
> Boston) wrote:
> 
> > I think it would be useful to consider how ambiguity (a/k/a URI  
> > collision)
> > http://www.w3.org/TR/webarch/#URI-collision
> > should be described in RDF.  For example, how would one 
> describe the  
> > meaning of the URIs http://fred.example/fluffy and 
> http://daphny.example/fido 
> >  in the following scenarios:
> >
> > 1. Fred owns http://fred.example/fluffy and declares that 
> "http://fred.example/fluffy 
> >  ambiguously denotes either my cat Fluffy or my web page that  
> > describes Fluffy".
> >
> > 2. Daphny owns http://daphny.example/fido and declares (perhaps  
> > using different communicateion mechanisms, or in different ways):
> > a. http://daphny.example/fido denotes my dog, whose name is Fido.
> > b. http://daphny.example/fido denotes a web page describing 
> my dog.   
> > Do you like my cool use of xhtml?
> >
> > ----------------------
> >
> > To my mind #1 and #2 are very similar, and the ambiguity can be  
> > modeled in N3 something like this:
> >
> > @prefix decl: <http://t-d-b.org?http://dbooth.org/2007/uri-decl/#> .
> > @prefix e: <http://example#> .
> > @prefix log: <http://www.w3.org/2000/10/swap/log#>.
> >
> > {
> >  {
> >  <http://fred.example/fluffy> a e:Cat .
> >  <http://fred.example/fluffy> e:hasName "Fluffy" .
> >  }
> > e:disjunction
> >  {
> >  <http://fred.example/fluffy> a e:WebPage .
> >  <http://fred.example/fluffy> e:describes "a cat named Fluffy" .
> >  } .
> > } decl:declares "http://fred.example/fluffy" .
> >
> > How would others express this in n3?
> >
> >
> >
> > David Booth, Ph.D.
> > HP Software
> > +1 617 629 8881 office  |  dbooth@hp.com
> > http://www.hp.com/go/software
> >
> > Statements made herein represent the views of the author 
> and do not  
> > necessarily represent the official views of HP unless 
> explicitly so  
> > stated.
> >
> >
> >> -----Original Message-----
> >> From: public-awwsw-request@w3.org
> >> [mailto:public-awwsw-request@w3.org] On Behalf Of Jonathan Rees
> >> Sent: Friday, January 30, 2009 1:15 PM
> >> To: public-awwsw@w3.org
> >> Subject: AWWSW telecon, Tues Feb 3
> >>
> >>
> >> Agenda as below. More use cases please -- or be prepared to declare
> >> there
> >> is no more semantics to be squeezed out of RFC 2616 and we
> >> should move
> >> on
> >> to another regime such as REST (i.e. what you want
> >> to impute to the sender if you're willing to believe they're using
> >> that regime).
> >>
> >> If you've forgotten what this is about see the top of
> >> http://esw.w3.org/topic/AwwswVocabulary
> >>
> >> Begin forwarded message:
> >>
> >>> From: Jonathan Rees <jar@creativecommons.org>
> >>> Date: January 17, 2009 9:34:32 AM EST
> >>> To: "public-awwsw@w3.org" <public-awwsw@w3.org>
> >>> Subject: AWWSW telecon, Tues Jan 20
> >>>
> >>> Agenda is continuing as before...  but I really really 
> would like to
> >>> find another use case where we can determine through 
> formal (or less
> >>> desirably, informal) reasoning that an HTTP response says 
> something
> >>> that is not true (or contradicts other intelligence). Ontology
> >>> building has to combine speculation with application, and
> >> we've been a
> >>> bit heavy on the speculation.
> >>>
> >>> Jonathan
> >>
> >>
> >>
> 



David Booth, Ph.D.
HP Software
+1 617 629 8881 office  |  dbooth@hp.com
http://www.hp.com/go/software

Statements made herein represent the views of the author and do not necessarily represent the official views of HP unless explicitly so stated.

Received on Monday, 2 February 2009 21:55:26 UTC