- From: Sandro Hawke <sandro@w3.org>
- Date: Mon, 09 Apr 2001 19:41:09 -0400
- To: Drew McDermott <drew.mcdermott@yale.edu>
- Cc: www-rdf-logic@w3.org
> Perhaps reification is just the ability to mark certain subexpressions
> as exempt from the usual conjunction flattening. If so, why not just
> declare *all* subexpressions exempt and forget reification?
Because RDF doesn't have subexpressions! (If I'm interpreting your
term correctly. RDF doesn't have parentheses. In the abstract
syntax, an RDF expression is a just a set of 3-tuples of atomic
identifiers.)
So when we want subexpressions, we use reification. It seems to work
fine, for the kind of things I've tried, where the logic is structured
in something like conjunctive normal form, so I can define which
terms are structural and which are subject to variable quantification.
I understand there's a big gap between RDF as defined in the published
recommendations and what you think is decent. I think there is less
of a gap between RDF as some of us are practicing it (and hope to it
to be defined in a newer spec soon) and what you want or think will
work.
-- sandro
Received on Monday, 9 April 2001 19:41:15 UTC