expressiveness and evolution issues for Rules Lite, some implications for syntax too

From: Benjamin Grosof (bgrosof@mit.edu)
Date: 09/30/03

  • Next message: Benjamin Grosof: "more thoughts on evolution of Rules Lite syntax"
    Hi folks,
    Here are some thoughts on expressiveness and evolution for Rules Lite.
    
    It's desirable as a design goal that we facilitate extensibility in the 
    direction of full Description Logic and FOL beyond that.
    
    In particular, it's vital that we facilitate Rules Lite rules being 
    combined expressively with OWL ontologies.
    
    It's vital as a design goal that we facilitate extensibility in the 
    direction of LP with nonmon and procedural attachments.
    
    An implication is the following:
    
    We should have the Rules Lite language be layerable on top of OWL, rather 
    than simply supersume OWL's expressiveness.
    This is because LP cannot handle all of even OWL-DL's 
    expressiveness.  E.g., the Rules Lite language can use properties/classes
    defined in OWL, but not have all of the expressiveness that OWL uses to 
    define complex classes.
    
    We should strive to specify the expressive constructs, and associated 
    syntactic constructs, of Rules Lite in terms that map cleanly/simply to 
    those of OWL.  Two important regards in which we need to add something are:
    - logical implication.  E.g., specify such via a new RDF property
    - logical variables.  E.g., specify such via a new RDF class
    
    Benjamin
    
    
    
    
    
    
    
    ________________________________________________________________________________________________
    Prof. Benjamin Grosof
    Web Technologies for E-Commerce, Business Policies, E-Contracting, Rules, 
    XML, Agents, Semantic Web Services
    MIT Sloan School of Management, Information Technology group
    http://ebusiness.mit.edu/bgrosof or http://www.mit.edu/~bgrosof
    


    This archive was generated by hypermail 2.1.4 : 09/30/03 EST