Re: A use case for anon nodes - action from telecon

"Aaron Swartz (by way of Ralph R. Swick )" wrote:
> 
snip
> 
> On Monday, July 23, 2001, at 11:02  PM, pat hayes wrote:
> 
> > I guess there is a genuine cultural clash here. Are we supposed
> > to define a precise meaning for RDF, or are we supposed to take
> > a breezy hack-it-up-if-we-need-it attitude to what RDF is all
> > about? I'm quite happy to take either road, but we need to get
> > the question clear, since the two attitudes arent really
> > compatible. The breezy approach has the merit of making
> > anything as precise as a model theory entirely pointless, so I
> > would have a lot less work to do.
> 
> Maybe we should keep it breezy for 1.0 and save the model theory
> for 2.0. I think it'll be sort of hard to retrofit semantics
> onto RDF when many folks haven't followed them (as we've seen).
> It also gives the whole group less work to do and a chance to
> spend time on issues that will affect the entire RDF community
> (as opposed to the logicians, etc. who are interested in a model
> theory).

Well, this is only indirectly about model theory.  I can understand how
the entire RDF community may not feel a compelling need for a thingie
[ahem!] titled "model theory", but how about a somewhat clearer RDF
M&S?  One, say, that clarified how anonymous nodes were to be
interpreted and handled.  You have to keep in mind that Pat's comment
was extracted from a message in a thread in which it was being suggested
that anonymous nodes could be used to represent variables in queries (or
templates).  Now, that may or may not be a reasonable thing to do, but
it is certainly out of the blue as far as my interpretation of RDF was
concerned.  I always thought RDF was intended for making assertions
about resources.  If I encountered some RDF on the Web, it was making
assertions about some resources.  Similarly, if someone *sent me* some
RDF, they were sending me some assertions about some resources.  I
wouldn't have interpreted that RDF as a query (and I'd hence have
misinterpreted any anonymous resources in that RDF if that were the
intent).  Now, as far as I can tell, the M&S today doesn't contain
anything that describes using RDF to represent queries and templates. 
That doesn't necessarily mean those uses are excluded or unreasonable,
but I claim that if we think that using RDF in these ways *is*
legitimate, it's our job to say so very clearly, and illustrate and
explain those uses in the M&S.  

-- 
Frank Manola                   The MITRE Corporation
202 Burlington Road, MS A345   Bedford, MA 01730-1420
mailto:fmanola@mitre.org       voice: 781-271-8147   FAX: 781-271-8752

Received on Tuesday, 24 July 2001 16:47:33 UTC