- From: Sandro Hawke <sandro@w3.org>
- Date: Thu, 07 Jul 2005 13:01:07 -0400
- To: Dave Reynolds <der@hplb.hpl.hp.com>
- Cc: public-rule-workshop-discuss@w3.org
> > So that's my quick analysis to get us started. What issues have I
> > missed?
>
> If I were dealing with Java objects or XML I think that analysis would be
> fine. However, in the case of supporting RDF data don't we also have to
> consider the semantics of the data, not just its format?
>
> If the rule language is going to be appropriate for RDF processing it
> probably ought to respect the semantics of things like bNodes (not to
> mention the open world assumption). Making a rule language based on case 3
> (XML as Fundamental Data) able to correctly process RDF data would surely
> take more than just an XPath-friendly RDF syntax.
>
> Or maybe that's an issue you'd rather separate off?
It does strike me as a separate issue. If the rule language itself is
a semantic extension of RDF (like SWRL is), then the challenge really
is just in dealing with the RDF/XML syntax, isn't it? The rule
language itself would have bNodes (file-scope existentially quantified
variables), etc. If the rule language has semantics which are
incompatible with deployed RDF and OWL, then yes, we have a different
set of challenges, well beyond the syntactic ones.
-- sandro
Received on Thursday, 7 July 2005 17:01:12 UTC