RE: Ideas for store for IFP smushing

> -----Original Message-----
> From: www-rdf-interest-request@w3.org
> [mailto:www-rdf-interest-request@w3.org]On Behalf Of ext John Black
> Sent: 16 August, 2004 17:40
> To: Stickler Patrick (Nokia-TP-MSW/Tampere); mof-rdf@mfd-consult.dk
> Cc: www-rdf-interest@w3.org
> Subject: RE: Ideas for store for IFP smushing
> 
> 
> 
> > From: Patrick.Stickler@nokia.com
> > Sent: Monday, August 16, 2004 2:52 AM
> > 
> > > -----Original Message-----
> > > From: ext Morten Frederiksen [mailto:mof-rdf@mfd-consult.dk]
> > > Sent: 15 August, 2004 21:00
> [snip]
> > > 
> > > Hi Patrick et al,
> > > 
> > > On Tuesday 10 August 2004 08:21, Patrick.Stickler@nokia.com wrote:
> > > > Note that the definition of a URIQA concise bounded 
> > > resource description
> > > > has been refined to be bounded by IFPs, to avoid "FOAF 
> > bloat" where
> > > > simple anonymous node closure produces an overly large 
> > description.
> > > > (c.f. http://swdev.nokia.com/uriqa/URIQA.html#cbd  for the 
> > > draft revision)
> > > Great move!
> 
> I believe concise bounded descriptions (CBD), in particular, and 
> URIQA in general are important and promising contributions to 
> making the semantic web. What is the status of efforts to make 
> this work into a standard? Was a note ever submitted to the W3C as 
> was suggested at one point? 

> How has URIQA been received? 

Folks seem either to love it or hate it. My own impressions after
2+ years of "debate" is that most folks involved in the historical
development and standardization of Web architecture have a NIH
bias against it, and have as yet been unnable to actually identify
any substantial flaw with the approach or offer a better altenative.

I.e., URIQA seems to be languishing in a political whirlpool and I
personally have not had the time/energy to "shmooze" it towards
a stronger position in the SW community.

I've decided to simply "lead by doing" and continue developing
code and proving its utility/benefit, and if anyone else cares
to take notice, they will.

Life's too short for politics (or, at least, my life is... ;-)

> What are 
> the objections raised? 

All objections raised thus far, such as they are, have been addressed 
in the FAQ section at the end of the URIQA specification. 

> I also have some questions (below) about the 
> definition of CBDs.
>
> [snip]
>  
> > How does the following definition work for you:
> > 
> > Given a URI denoting some resource, a "concise bounded 
> > description" of that 
> > resource is a set of RDF statements, explicitly asserted 
> 
> How is this "explicitly asserted" condition determined? 
> I mean what counts as an "explicitly asserted RDF statement"? 
> as opposed to a merely "asserted RDF statement"? or just a 
> plain, let us say, "published RDF statement"?

Explicitly asserted simply means that a triple exists in the
graph, as opposed to a triple that might be inferable from
some other triple(s) or from some rule.

I.e., a URIQA implementation is free to employ inference (or not)
when determining what statements should be included in a CBD.

In the Nokia implementation, there is a parameter which controls
whether inference is employed. 

Arguably, one could leave any reference to inference out of 
the CBD definition, and leave it up to the agent to derive
one graph from another via inference and speak of extracting
the CBD from the derived graph.

Historically, it seemed simpler to present it as I have, but
it may warrant some editorial attention.

> 
> > and/or inferred, 
> > comprised of the following:
> > 
> >    1. Include all statements 
> 
> What is the intended scope of the "all" here? Possiblities 
> include "all statements with the same namespace", or "all statements 
> with this domain name", or "all statements anywhere on the web"? 

All statements contained in the graph (knowledge base) from which the 
CBD is extracted.

I.e. the URIQA spec is written from the perspective of a SW agent
which manages a certain body of knowledge, and from which one can
ask for descriptions of resources insofar as that agent has
knowledge about such resources.

There would be one and only one authoritative agent, which can
be queried using the URIQA HTTP methods -- yet there can also be any
number of non-authoritative agents, which can be queried using the
URIQA web service API. In either case, though, the query for a
CBD of some resource is within the context of a body of knowledge
maintained by each particular agent -- and a queries submitted to
several agents about the same resource can be syndicated into a
single unified description about that resource.

Thanks for the questions, John. I'll think about how I can clarify
the spec.

Cheers,

Patrick

--

Patrick Stickler
Nokia, Finland
patrick.stickler@nokia.com

Received on Tuesday, 17 August 2004 05:49:12 UTC