Re: Proposed text for Choreography concepts

Hi Mike.

It looks good to me. A few comments:

* Champion, Mike <Mike.Champion@SoftwareAG-USA.com> [2003-06-25 19:49-0600]
> Compare with http://www.w3.org/TR/2003/WD-ws-arch-20030514/#choreography
> 
> Executive summary: wordsmithed the draft text a bit to be more in line with
> recent discussions in the Choreography WG relating "choreography" to a
> shared global state machine rather than an execution language.
> Cross-checked language and issues here against the Choreography WG charter
> and various email-threads, notably:
> http://lists.w3.org/Archives/Public/www-ws-arch/2002Aug/0101.html
> http://lists.w3.org/Archives/Public/www-ws-arch/2002Aug/0191.html
> http://lists.w3.org/Archives/Public/www-ws-arch/2002Aug/0193.html
> http://lists.w3.org/Archives/Public/www-ws-arch/2002Oct/0205.html
> http://lists.w3.org/Archives/Public/www-ws-arch/2002Oct/0369.html
> http://lists.w3.org/Archives/Public/www-ws-arch/2003May/0033.html
> 
> Issues:
> 
> - Is the "shared state machine" definition too idiosyncratic and specific to
> the W3C WS-Chor WG? 

The question which is inevitabily going to come up though is going to
be what about BPEL-like languages, both for abstract and executable
processes, and the concepts they describe.

So I think that your definition is fine, but we are going to need to
address this other type of description.

> - Doe we want to wade into the swamp of defining "orchestration" or should
> we follow Martin's lead and simply ban the term :-)  If so, does it mean
> "choreography implementation" or "sortof like choreography, but from a
> particular actor's point of view?  See
> http://lists.w3.org/Archives/Public/www-ws-arch/2003May/0034.html

Hmmm... being offline, I can't see this message unfortunately, but I
think that this is related to my point above, with managing to put a
label on those concepts.

> - What's the relationship between Choreography and MEP?  There was a
> sentiment at the WS-Chor F2F last week that a Choreography language should
> be able to model any MEP.

There was some good text proposed by Mark Jones and linked from our
document about that:

  http://lists.w3.org/Archives/Public/www-ws-arch/2003Jan/0494.html

Now may be a good time to incorporate it and try to nail down this MEP
beast.

[..]
> YYY Choreography Description Language

A question about this one. I remember agreeing in Rennes that
languages should not be considered as concepts.

> YYY.1 Definition
> 
> A Choreography Description Language is a notation for describing a
> choreography.  It may also permit the specification of a composite service
> in terms of component services.
> 
> YYY.2 Relationships to other elements
> A choreography Description Language describes
> the pattern of allowable interactions between a set of services
> 
> A choreography Description Language may describe
> the life cycle of a service invocation
> 
> A choreography Description Language describes
> the conversations possible between service requesters and service providers.

A choreography description language describes three things listed
above, whereas a choreography was just the first one of them. Is that
intended?

Conceptually, I would have imagined a foo language describing foo, no
more and no less.

Regards,

Hugo

-- 
Hugo Haas - W3C
mailto:hugo@w3.org - http://www.w3.org/People/Hugo/

Received on Thursday, 26 June 2003 10:11:56 UTC