RE: Definition of Choreography

Another way to look at the distinction, if it is useful:  is the behavior of the system "scripted" in some direct (procedural/imperative) way, or "emergent" from rule/condition interactions?

Scott

At 4:34 PM -0400 10/17/02, Mathews, Walden wrote:
>'Determinate' might be a good word there, instead of
>or in addition to 'turing complete'.
>
>WM
>
>> -----Original Message-----
>> From: David Orchard [mailto:dorchard@bea.com]
>> Sent: Thursday, October 17, 2002 4:26 PM
>> To: 'Mark Baker'; 'Champion, Mike'
>> Cc: www-ws-arch@w3.org
>> Subject: RE: Definition of Choreography
>>
>>
>>
>> When i said Turing complete, I mean distinguising between two
>> styles of
>> choreography.  A choreography can specify an order between two nodes.
>> Imagine that a node sends Messages M1 or M2 depending upon
>> some particular
>> variable.  In a non-turing complete, the choreography would
>> say M1 or M2.  A
>> turing complete choreography language would say something
>> like If C1 then
>> send M1 else send M2.
>>
>> Cheers,
>> Dave
>>
>> > -----Original Message-----
>> > From: www-ws-arch-request@w3.org
>> [mailto:www-ws-arch-request@w3.org]On
>> > Behalf Of Mark Baker
>> > Sent: Thursday, October 17, 2002 6:06 AM
>> > To: Champion, Mike
>> > Cc: www-ws-arch@w3.org
>> > Subject: Re: Definition of Choreography
>> >
>> >
>> >
>> > On Wed, Oct 16, 2002 at 09:50:12PM -0400, Champion, Mike wrote:
>> > > I think of "Choreography" sortof like a policy, not a program.
>> >
>> > I agree.
>> >
>> > But David said something that suggested that it was defining
>> > the *how*,
>> > not just the *what*; "specification of ordering of messages".  If it
>> > were to define the *what*, I would expect it to say something like;
>> > "The specification of potential state changes".
>> >
>> > In most cases, there are multiple possible sequences of messages
>> > that could result in a desired state change.  As a trivial example,
>> > any sequence that included an HTTP GET message, could include an
>> > arbitrary number of HTTP GETs.  i.e. POST-GET-POST is equivalent to
>> > POST-GET-GET-GET-GET-POST.
>> >
>> > Also, the mention of turing completeness suggests *how*, rather than
>> > *what*, though I'm a bit unclear about its intent due to the
>> > use of the
>> > term "message exchange pattern" (which presumably means something
>> > different than a SOAP MEP - perhaps "message exchange sequence"?)
>> >
>> > On the plus side, I like that it's short. 8-)
>> >
>> > MB
>> > --
>> > Mark Baker, CTO, Idokorro Mobile.  Ottawa, Ontario, CANADA.
> > > http://www.markbaker.ca             http://www.idokorro.com
> > >
> > >
> >


-- 
Scott Vorthmann                    mailto:scottv@tibco.com
Senior Architect                     mailto:scottv1@imcingular.com
                                              office: 919 969 6513
TIBCO Extensibility                  mobile: 919 593 2349
TIBCO Software, Inc.               http://www.tibco.com

Received on Thursday, 17 October 2002 16:59:06 UTC