ACTION-170: List (clarify) issues raised by the extensibility requirement

This is an incomplete list of issues/approaches related to the
extensibility requirement. In particular, we want to ensure forward
compatibility of the RIF Core with dialects. The items are listed in no
particular order.

 

*	asn06 notation (and OWL) allows us to extend the language fairly
easily which is good for defining dialects. For example, a reaction rule
including a post-condition could be easily defined.
*	Sandro Hawke's proposal (summarized by Dave Reynolds):
"translators can (should) implicitly of explicitly resolve extension
syntax elements to fetch the associated syntax extensions and "must
understand" metadata."
*	Standard library of building blocks including

	*	math relations;
	*	constants;
	*	numbers;
	*	functions;
	*	standard built-ins including those used in expected
common dialects (for example, assert);
	*	etc.

*	Propose an approach for rules consisting of multiple parts. Body
and head notation is not very meaningful for PR or RR. Metadata (RDF
scaffolding) is one approach, asn06 inheritance is another.
*	As pointed out by Gerd Wagner and also mentioned in the e-mail
on JBoss Rules RIF mapping, F-Logic mapping requires typing and possibly
vocabulary facts to be expressible.
*	Dave Reynolds proposal for metadata as an extensibility
mechanism
(http://lists.w3.org/Archives/Public/public-rif-wg/2006Sep/0077.html).
Metadata for rules modules, priorities, namespace mapping, ordering
(import for guaranteed round-tripping or operational semantics in serial
Horn parts). The idea is to exploit RDF extensibility to deal with
various extension issues including the rule parts in dialects. We have
already implemented this in the recent Prova version and it works very
nicely allowing the rules to reason about metadata.
*	In PR, actions in the action part are _not_ interchangeable (for
example, retract and assert), as well as the condition part may have
different behaviour depending on order. The same may apply to rule body
in some dialects of reaction rules.
*	Using CSP as a mechanism for separating and constraining the
data model is also an extensibility mechanism.
*	H2F extensions suitable for extensions to distributed logic
programming and reactive systems.

 

 

Alex Kozlenkov

Advanced Technology Group

 

Office: +44 20 8834 6854

Yahoo! Messenger: alex.kozlenkov

 

Betfair Limited | Winslow Road | Hammersmith Embankment | London | W6
9HP

Company No. 5140986

 

The information in this e-mail and any attachment is confidential and is
intended only for the named recipient(s). The e-mail may not be
disclosed or used by any person other than the addressee, nor may it be
copied in any way. If you are not a named recipient please notify the
sender immediately and delete any copies of this message. Any
unauthorized copying, disclosure or distribution of the material in this
e-mail is strictly forbidden. Any view or opinions presented are solely
those of the author and do not necessarily represent those of the
company.