W3C home > Mailing lists > Public > www-rdf-rules@w3.org > November 2003

Re: Rules WG -- draft charter -- NAF

From: Andrei Lopatenko <lopatenko@inf.unibz.it>
Date: Mon, 17 Nov 2003 22:49:14 +0100
Message-ID: <3FB9425A.9040703@inf.unibz.it>
To: Drew McDermott <drew.mcdermott@yale.edu>
Cc: www-rdf-rules@w3.org

I believe there are some other reasons (important?)  to clearly define a 
scope for statements at the web nesides reasons mentioned before 
(imported statements, CWA reasoning).
At some point in future, when more and more sophisticated logics will be 
used for SW reasoning, it will be possible that derivation of negative 
information will be supported in SW what may lead to inconsinstent 
theories..
(it is not supported now by OWL Rules, but negative information maybe 
derived by OWL reasoning itself, if regard ontologies in the same way as 
we treat RDf data (distributed, CW etc))
If inconsistency is derived in some theory (agent) what the boundaries 
for it, should it be restricted to soem space or make all (dependent) 
web inconsistent.
I believe a framework to restrict inconsistency should be specified.
Maybe it is not actual and important now  with onlu positive heads in 
OWL rules , but probably issues hsould be considered already
Andrei Lopatenko
PhD student
The University of Manchester
RA
Free University of Bozen-Bolzano


Drew McDermott wrote:

>  [Ben Grosof]
>   I know that some don't like the idea of having to do this.  I think the 
>   alternative of not being allowed to define such scoping is, however, 
>   extremely undesirable.  The idea of "all RDF anywhere on the web" as 
>   something I would want to always *have to* use as my KB's scope is a 
>   complete non-starter practically -- consider issues of data/knowledge 
>   quality alone!  (I'm tempted to say it's ridiculous.  People talk about 
>   "trust" on the Semantic Web.  The most basic mechanism for trust is simply 
>   to know what set of premises the inferences were drawn from.  We'll be 
>   laughed out of town in most practical IT settings if we don't have a good 
>   story about this aspect of things.)
>
>   ...
>
>I have little to add here, except to agree with Ben 100%.  Jim's
>example sounds superficially plausible, until you realize that if you
>must follow one innocent-looking pointer to a dataset, then by
>transitivity you must take the entire web into account, which is
>absurd.
>
>As a corollary, the mere use of a namespace doesn't imply that an agent
>must examine (let alone accept) the contents of some document
>associated with that namespace.  It's as if calling someone's theory
>_merde_ requires accepting everything published in Figaro.
>
>  
>
Received on Monday, 17 November 2003 16:45:43 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 2 March 2016 11:10:15 UTC