W3C home > Mailing lists > Public > public-rif-comments@w3.org > June 2012

Semantics of Action Block with Variable Declarations in RIF-PRD

From: Jesse Weaver <weavej3@rpi.edu>
Date: Tue, 5 Jun 2012 16:46:32 -0400
Message-Id: <0D271438-BC49-414B-8D01-95C5687FCA9E@rpi.edu>
To: public-rif-comments@w3.org
I have a question about the semantics of action blocks with action  
variable declarations.  Suppose an action block has an action variable  
declaration of the form (?oldValue ?shoppingCart[ex1:value->?oldValue)  
where ?shoppingCart is bound to some constant -- say ex1:sc -- from  
matching the rule condition formula (this is actually part of example  
3.2 in the RIF-PRD recommendation).  What happens if the set of facts  
representing the state of the fact base does not contain any ground,  
frame atomic formula with object ex1:sc and attribute ex1:value (or  
put another way, the is no constant c such that ex1:sc[ex1:value->c]  
matches the set of facts)?  Does the rule just fail to fire?  That  
seems like the intuitively correct behavior since the frame must be  
matched to the set of facts according to the definition of action  
instance.  However, I cannot find an explicit specification in the  
recommendation concerning the appropriate behavior for this case.   
Everything seems to assume that the frame matches the set of facts.

Jesse Weaver
Ph.D. Student, Patroon Fellow
Tetherless World Constellation
Rensselaer Polytechnic Institute
http://www.cs.rpi.edu/~weavej3/index.xhtml
Received on Tuesday, 5 June 2012 20:49:48 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 5 June 2012 20:49:48 GMT