Re: Object vs fact <-- Reference vs import <-- RIF Core shortened

s/non normative/normative  and that pretty much sums it up.

and I find it an odd notion that we could somehow have # and ## as facts 
in the abstract syntax and semantics in order to define what import 
means but we wouldn't expose these in the concrete syntax. 

Paul Vincent wrote:
> My interpretation of Gary's position is:
>
> "PRD (and possibly RIF core) must include (non normative) definitions
> for mappings from common XSD, POJO, and possibly RDBMS object and data
> structures and types to fact models and vice versa." 
>
> In fact, just like the BLD team have done for OWL and RDF...
>
> Paul Vincent
> TIBCO | Business Optimization | Business Rules & CEP
>  
>   
>> -----Original Message-----
>> From: Gary Hallmark [mailto:gary.hallmark@oracle.com]
>> Sent: 21 November 2008 07:35
>> To: kifer@cs.sunysb.edu
>> Cc: Christian de Sainte Marie; Paul Vincent; Patrick Albert; Dave
>> Reynolds; Boley, Harold; Adrian Paschke; Axel Polleres; RIF WG
>> Subject: Re: Reference vs import <-- RIF Core shortened
>>
>> My view is that the phrase "external object model" is a misleading .
>>     
> It
>   
>> is more accurate to think of importing an XML document or a java
>>     
> object
>   
>> graph.  Import defines a mapping between the external syntax and the
>>     
> RIF
>   
>> syntax.  With a standard mapping and a standard RIF semantics you get
>>     
> a
>   
>> standard semantics for importing the external syntax.  We can't map
>> everything from an XML document or a java object graph into Core but
>>     
> at
>   
>> least we can map objects with slots, membership, and subclass.
>>
>> Put another way: if you want to exchange rules using a fancy external
>> data model, you need to figure out how to map it to your RIF dialect
>>     
> in
>   
>> a standard way.
>>
>> Michael Kifer wrote:
>>     
>>> ok. But I am trying to make this more concrete so that we'll
>>>       
> understand.
>   
>>> I would like things to be expressed in the context of RIF-Core and
>>>       
> of
>   
>> the
>>     
>>> concrete problem for which Gary was seeking a solution.
>>>
>>> I don't see how Java objects and external schemas relate to allowing
>>>       
> #
>   
>> and ##
>>     
>>> in RIF-Core facts.  I would like to see a clarification from you on
>>>       
> that
>   
>> issue
>>     
>>> and Gary's view.
>>>
>>> michael
>>>
>>>
>>> On Thu, 20 Nov 2008 16:22:50 +0100
>>> Christian de Sainte Marie <csma@ilog.fr> wrote:
>>>
>>>
>>>       
>>>> Paul Vincent wrote:
>>>>
>>>>         
>>>>> Christian's comment is simply (?) that RIF needs to play well
>>>>>           
>> alongside
>>     
>>>>> externally-defined fact definitions (for example external Java
>>>>>           
> object
>   
>>>>> models used to define production rules in BREs).
>>>>>
>>>>>           
>>>> Thanx for translating from the csma-ese, Paul :-)
>>>>
>>>>
>>>>         
>>>>> Maybe the qu is whether it is compulsory that all relevant facts
>>>>>           
> and
>   
>>>>> class relationships need to be represented in RIF for RIF rules to
>>>>>           
> be
>   
>>>>> defined against them?
>>>>>
>>>>>           
>>>> It is compulsory that they need be representable; so, yes, they
>>>>         
> could
>   
>> be represented.
>>     
>>>> But it is not compulsory that they be represented, as far as I
>>>>         
>> understand.
>>     
>>>> That is, by the way, what I understand Gary says in his reply to
>>>>         
> you
>   
>> [1], and this is, anyway, what I have been trying to say all along.
>>     
>>>>         
>>>>> Or have I missed the point (again)? :)
>>>>>
>>>>>           
>>>> I do not think so.
>>>>
>>>> Cheers,
>>>>
>>>> Christian
>>>>
>>>> [1]
>>>>         
> http://lists.w3.org/Archives/Public/public-rif-wg/2008Nov/0127.html
>   
>>>>
>>>>         
>>>       
>
>

Received on Friday, 21 November 2008 19:54:40 UTC