RE: [sca-assembly] [SCA EVENTS] Summary of tentative agreements

["Martin Chapman" <martin.chapman@oracle.com>]

Scott,

Can we at least agree it's a point for discussion since there are differing views here?

Other comments below.

Martin.

> Subject: Re: [sca-assembly] [SCA EVENTS] Summary of tentative agreements
> 
> Martin,
> 
> I was afraid that I mis-heard Peter's question to me, and it looks
> like I did.  I'm not in agreement with number 4, and I think I can
> safely say I speak for TIBCO on this.
> 
> In my view, the constraints on "wiring" and interface compatability
> should be contingent upon the character of the interface.  For
> example, an interface entirely full of one-way operations (I'll call
> them events) does not require a producer of those events to produce
> any or all of them; it merely guarantees that no *other* events will
> be produced by that producer.

[<MartinC>] I think you are going to have to clarify "who" the interface is defined by.
However this seems to me like saying a client specifying a reference is not obliged to invoked any operation on the service, which
in my mind is not worth saying;) As to the other point, can a producer produce things that it didn't declare, we had a long
discussion on that in OSAO and compromised on a SHOULD - one of the rationale being that in wsdl you can define faults, but other
faults may be sent that were not explicitly declared (e.g. SOAP faults)

> 
> Our current wiring and interface compatibility rules simply make the
> assumption that every interface has at least one request-reply
> operation.

[<MartinC>] Mike E replied to that point, but another discussion point is would/can a one-way operation mixed in with
request-responses in an interface be considered as an event or not i.e. what is special about that one-way in a mixed interface vs
an interface exclusively comprised of one-ways? I think questions like this are interesting, and point to the fact that while we
might not agree about position 4, we should explore the area?

> 
> Scott
> 
> On Sep 10, 2009, at 6:22 AM, Martin Chapman wrote:
> 
> > Here is a summary of what I think we agreed yesterday, and are
> > therefore areas for the TC to explore further.
> >
> >
> >
> > Position 1: the current SCA 1.1 model needs tweaking/adding to, to
> > support a pub/sub paradigm.
> >
> >
> >
> > Position 2: SCA 1.1 use of WSDL has certain assumptions in
> > assemblers'/developer's heads (and current tooling), which need  re-
> > examining for pub/sub.
> >
> >
> >
> > Position 3: SCA 1.1 wires and wiring may be too restrictive for pub/
> > sub and need to be relaxed/extended.
> >
> >
> >
> > Position 4: We agreed that an assembler looking at SCDL should be
> > able to distinguish between events and one-way requests.
> >
> >
> >
> > Cheers,
> >
> >    Martin.
> >
> >
> >
> >
> > Martin Chapman | Standards Professional
> > Mobile: +353 87 687 6654
> >
> > ORACLE Ireland
> > "Please consider your environmental responsibility before printing
> > this e-mail"
> >
> 
> 
> ---------------------------------------------------------------------
> To unsubscribe from this mail list, you must leave the OASIS TC that
> generates this mail.  Follow this link to all your TCs in OASIS at:
> https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php