RE: problems with concise bounded descriptions

> -----Original Message-----
> From: www-rdf-interest-request@w3.org
> [mailto:www-rdf-interest-request@w3.org]On Behalf Of ext John Black
> Sent: 01 October, 2004 16:43
> To: Peter F. Patel-Schneider; www-rdf-interest@w3.org
> Subject: RE: problems with concise bounded descriptions
> 
> 
> 
> The following represents my understanding of the note. As such it 
> may point out both ways in which the note could be clarified or 
> ways my understanding is deficient. 

And I continue to be very open and welcome input on
ways the document can be improved.

> Also, given the genesis of CDB 
> as the result of a URIQA query on a URIRef, some of what I 
> say recalls 
> that history. Peter points out additional uses that could be 
> made of CDBs 
> by generalizing the query to include blank nodes as the 
> input, 

The present "presentation" (I'll avoid the use of the word "definition"
as it seems to be problemmatic) of CBD allows for extracting CBDs
of blank node subjects.

How you get to a particular node in a particular graph, from which
you extract the CBD is IMO an query interface/API issue.

> but its 
> first use case, as I understand it, was to provide a bounded set of 
> descriptive expressions constraining the interpretation of a 
> single URIRef 
> taken in isolation. In short, its first use was as part of a naming 
> performative consisting of the four parts:
> 
>  1. A naming authority - the source of both the URIRef and the CDB
>  2. A URIRef used as a name of a resource (thing in the world)
>  3. A graph constraining the possible interpretations of that 
>     URIRef to one unique thing in the world.
>  4. A public announcement (publication on the web) by the source 
>     that the object described by the graph is the referent of 
> the name.

Essentially, yes. And that remains its primary use case in most
applications where we use CBDs, but CBDs are not intended to be
constrained to that particular use case.

> 
> > -----Original Message-----
> > From: www-rdf-interest-request@w3.org
> > [mailto:www-rdf-interest-request@w3.org]On Behalf Of Peter F.
> > Patel-Schneider
> > Sent: Thursday, September 30, 2004 7:40 PM
> > To: www-rdf-interest@w3.org
> > Subject: problems with concise bounded descriptions
> > 
> > 
> > 
> > In the DAWG message archive I came across a reference to a 
> W3C member
> > submission from Nokia on Concise Bounded Descriptions
> > http://www.w3.org/Submission/CBD/.
> > 
> > The notion of Concise Bounded Descriptions (CBD) in this note 
> > has a number
> > of problems.
> > 
> > The initial description of a CBD is severely underspecified.  
> > According to
> > the note, ``A [CBD] of a resource is a body of knowledge about that
> > resource which does not include any explicit knowledge 
> about any other
> > resource which can be obtained separately from the same source.''
> > 
> > Problem 1:  Which source?
> 
> The source is the agent with the authority created by possessing the 
> means to legitimately implement a web response to the URIRef 
> in question;
> the agent with such authority that is motivated to promote 
> through public 
> announcement a common knowledge of a mechanical substitution of that 
> URI-used-as-symbol for a graph containing additional expressions 
> in description of the denotation of that URIRef. In terms from the 
> RDF Semantics spec., the agent here is publishing expressions 
> that constrain the mapping IS from the URIRef as a vocabulary term 
> to that object in the real world that the source agent is mapping 
> it to when the source agent uses it.

This discussion about "source" perhaps brings into play a bit
too much overshadowing of the applications of CBDs (the query
interfaces and inter-agent communication) than is necessary.

I will think about how it can be distilled and generalized further.

(historically, the definition of CBDs was a component of the URIQA
specification, and of course, still remains central to URIQA, and 
surely still contains influences from that prior context).

Patrick

Received on Friday, 1 October 2004 15:29:03 UTC