W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > June 2003

recent tweaks to RDF semantics

From: pat hayes <phayes@ihmc.us>
Date: Tue, 10 Jun 2003 20:50:24 -0500
Message-Id: <p0521060abb0c37178eed@[10.0.100.24]>
To: w3c-rdfcore-wg@w3.org
Cc: Brian_McBride <bwm@hplb.hpl.hp.com>

  http://www.ihmc.us/users/phayes/RDF_Semantics_Edit_nT.html

now has some tweaks which fix some bugs noted in recent emails.

The 'se' rules (now moved into section 4 which now has all the rule 
stuff in one place) have been modified so that they only generate a 
single bnode from each URiref or literal in the original graph. This 
is enough for the closure lemmas and keeps them from generating all 
those redundant blank-node copies.

The definitions of rdf and rdfs interpretations have been slightly 
tweaked so that the relevant vocabularies contain the rdfV and rdfsV 
vocabularies (previously, an rdf-interp of V was an interpretation of 
V+rdfV; now it is just of V, so one has to be explicit about V 
including the 'special' vocabulary; similarly for rdfsV).  The only 
effect of this tweak (suggested by Dan C) is to make it possible for 
an rdf-interpretation to only interpret the part of the infinite 
container vocabulary that it actually needs; so the definitions of 
closure are now relativized to a vocabulary, and now refer to those 
axioms which contain names from that vocabulary; and the closure 
lemmas now refer to closures relative to the vocabulary of (S+E).

I have rewritten rule rdf2 so that it replaces an XML literal by its 
canonical form, rather than allowing arbitrary equality 
substitutions. The exact wording may need massaging, but the idea 
seems sound.

Taken together, all this enables us to completely avoid the 
'infinite' cases in closures, so all closures (from finite graphs wrt 
a finite vocabulary) are finite, without any need of weasel-words.

The weird rule rdfs12 has been further modified. The elegant version 
suggested by Jos unfortunately turned out to not be quite valid. This 
rule may still get some further tweaking but this will not change any 
other part of the MT.

The Lbase appendix is now in line with the rest of the document; the 
datatyping in particular has been rationalized.

Apart from getting the XML literal terminology exactly aligned with 
Concepts, and writing out the proofs properly, I believe this is now 
pretty much done.  On the other hand, Ive said that before....

Pat

-- 
---------------------------------------------------------------------
IHMC	(850)434 8903 or (650)494 3973   home
40 South Alcaniz St.	(850)202 4416   office
Pensacola			(850)202 4440   fax
FL 32501			(850)291 0667    cell
phayes@ihmc.us       http://www.ihmc.us/users/phayes
Received on Tuesday, 10 June 2003 21:50:30 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:57:54 EDT