Outcomes from Jan 2 telecon

Although we came to no resolutions today, I remain optimistic that we 
are close to having something for the RIF Core.  I want again to 
summarize what I think happened today to be sure I understood it, 
especially the cases where I thought there was agreement.

The main technical discussion at today's telecon centered again on the 
idea of "slots."

Harold agreed that the positional -> keyword mapping with predicate 
signatures would address his concerns.

Michael however pointed out that there are at least three semantics in 
use for "slots": relational, psi-terms, and the (nameless?) semantics 
used by F-Logic.  He is working on a document that clarifies what 
these semantics are, which will be out within the next day or so.

Hassan wasn't sure at first that this was any problem, suggesting that 
we keep one syntax and let the constraints carry these semantics. 
Several people felt that the differences should be reflected in the 
syntax as well, i.e. a different syntax for each "kind" of slot. 
Hassan disagreed, but Michael claimed to have some use cases from 
users of F-Logic that they might want a single rule set that supported 
more than one of these semantics.  Hassan then seemed to agree that if 
the different "kinds" of slots were to be mixed in a single ruleset, 
they should in fact be identified in syntax.

-Chris

-- 
Dr. Christopher A. Welty                    IBM Watson Research Center
+1.914.784.7055                             19 Skyline Dr.
cawelty@gmail.com                           Hawthorne, NY 10532
http://www.research.ibm.com/people/w/welty

Received on Tuesday, 2 January 2007 19:48:21 UTC