[closed] timbl-02, reification semantics


With reference to:

The RDFcore working group have considered your comment about changing the 
semantics of reification, and subsequent suggestions to remove reification 
altogether [2], and has resolved:


not to accept this comment.  (This response does not address the bagId 
question you raised, which is being dealt with in a separate response.)

In our discussions, we noted three significant applications that use 
reificiation as currently defined ([1], [5], and also Roland Schwaenzl 
(http://www.mathematik.uni-osnabrueck.de/staff/phpages/schwaenzlr.html) was 
reported to use reification as currently defined).  We note and agree that 
reification as defined does not address the particular problems you hoped 
it would solve [3], but considering that other users of RDF have found it 
useful leads us to the conclusion that reification should remain as defined.

We believe that the clarification of reification that you suggest [3] is 
provided by the discussion and non-entailment mentioned in the RDF formal 
semantics [4].

The working group has agreed to include some additional "health warnings" 
on the corresponding semantics and schema descriptions so that folks are 
alerted to some uses for which reification is not appropriate.

Can you please respond to <www-rdf-comments@w3.org> indicating whether this 
response is satisfactory.  Thank you.

[1] http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2003Apr/0108.html

[2] http://lists.w3.org/Archives/Public/www-rdf-comments/2003JanMar/0241.html

[3] http://lists.w3.org/Archives/Public/www-rdf-comments/2003JanMar/0477.html
(recorded as issue timbl-02:)

[4] http://www.w3.org/TR/rdf-mt/#Reif

[5] http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2003Apr/0167.html


Graham Klyne
PGP: 0FAA 69FF C083 000B A2E9  A131 01B9 1C7A DBCA CB5E

Received on Wednesday, 30 April 2003 04:43:36 UTC