RE: soundness and semantics

 
(see below for comments)

-----Original Message-----
From: public-rif-wg-request@w3.org [mailto:public-rif-wg-request@w3.org] On
Behalf Of Sandro Hawke
Sent: Thursday, May 11, 2006 2:44 PM
To: Francois Bry
Cc: Chris Welty; public-rif-wg@w3.org
Subject: Re: soundness and semantics 


   ... (deleted) ...
    
    > 2. Such an interchange is only possible if the original semantics of 
    > the rules and rule sets to be interchanged is specified. Therefore, 
    > the possibility to specify a semantics for a rule or rule set should 
    > be a requirement.
    
    That only comes up if there were rulesets with the same syntax by
different semantics, right?  But that's only one approach: we could use
different syntax (eg different namespaces or different element names)
whereever we need different semantics.
    
    I wrote about this in the notes from F2F1:
    
    http://www.w3.org/2005/rules/wg/wiki/Semantics_Extensibility_Point  

[stan]  
This, by the way, is almost the approach taken in MathML.  Re-use of the
basic functional syntactic structure was essential from the beginning so the
"definitionURL" (really a URI) attribute was explicitly required whenever
the "Default semantics" were violated.  The definitionURL attribute
qualified the mathematical meaning, much in the spirit of xmlns, but in a
way that allowed re-use of the underlying syntax and the syntactic
validation mechanisms.

I suspect all our RIF dialects can also use a common syntax (to be
identified) and that a similar mechanism could flag divergence from a
default meaning.

To my mind the reasons we did not go to simple reliance on a pure namespace
based distinction included a strong desire for a static syntax. 

- This helped to ease validation, distribution of documents, etc., and the
mathematical extensions come in how you use that notation. 

- It made it possible to identify the syntactic role of an object and with
that some understanding of it, e.g. that it was a function) without knowing
anything about it beyond that.  

- Once you switched namespaces, all bets as to meaning were off and the
machinery to extend the structure in a way that recaptured or preserved some
of that syntactic structure, while available, was much more sophisticated
than just choosing a URI.

There was also a relative absence of namespace mature applications at that
time and we also wanted wide distribution :) 

(The missing piece is the algebra of dialects.)

[stan]

Received on Thursday, 11 May 2006 13:46:53 UTC