RE: [PRD] Issues to resolve before publication (Execute)

In PRR we use Execute as a catch-all. Given we are not going to
implement the entire universe of DSLs (including associated methods) in
RIF, by definition, we will need an external / execute. 

But I have no opinion as to whether this appears in the PRD draft or
later...

Ref: DSL = Domain Specific Language, meaning usually something like a
domain-specific schema like MISMO, ACORD, RosettaNet, XBRL, etc etc

Paul Vincent
TIBCO | Business Optimization | Business Rules & CEP
 
> -----Original Message-----
> From: public-rif-wg-request@w3.org
[mailto:public-rif-wg-request@w3.org]
> On Behalf Of Christian de Sainte Marie
> Sent: 01 July 2008 12:42
> To: Gary Hallmark
> Cc: RIF WG
> Subject: Re: [PRD] Issues to resolve before publication (Execute)
> 
> 
> Gary Hallmark wrote:
> >
> > No, the definition of "external" is that its semantics is fixed and
> > independent of any rules that call it.  So if I send you a ruleset
with
> > a reference to pred:numeric-equal in it, you know the semantics
thanks
> > to DTB.  But if I send you a ruleset with a reference to
> > gary:secret-function in it, you have no idea.
> 
> Yes, unless we share the specification of gary:secret-function out of
> band (like for the pred: and fun: externals, the spec of which is
shared
> in the out-of-band DTB document).
> 
> If we want PRD to be used to interchange existing rules (yes: legacy),
> we need this kind of externals (I mean: application/user-specific).
> 
> If we want them, they can only be dealt with as balckboxes, as far as
I
> can think: their semantics will thus be the same as what is in the
> current draft, in one form or another.
> 
> Which is why I thought that it made sense to include Execute.
> 
> But I may be wrong, of course: I have been enlightened yet, as you all
> know :-)
> 
> Cheers,
> 
> Christian
>

Received on Tuesday, 1 July 2008 12:17:11 UTC