W3C home > Mailing lists > Public > public-ws-chor@w3.org > July 2003

RE: Revised: Mission Statement

From: Cummins, Fred A <fred.cummins@eds.com>
Date: Wed, 2 Jul 2003 15:11:35 -0500
Message-ID: <1A254DC4B97D8C4CB4A5611CF8058F5F01080E6C@USPLM214>
To: Nickolas Kavantzas <nickolas.kavantzas@oracle.com>, jdart@tibco.com
Cc: Martin Chapman <martin.chapman@oracle.com>, Steve Ross-Talbot <steve@enigmatec.net>, "Yaron Y. Goland" <ygoland@bea.com>, public-ws-chor@w3.org
Nickolas,
 
It may be a neutral view model, but it does not describe "common" behavior.
It describes the complementary behavior(s) of the participants, and it
defines the "steps" by which a "business transaction" may be achieved (the
preferred or mainstream path).  Consequently, the message sends and receives
of the individual participants represent the complementary behavior(s).  The
"wiring" between them defines the paths over which messages travel, and this
the behavior observable by each.
 
Also note that the model could include exchanges with other participants
that are not part of the primary exchange, e.g., my example of the buyer,
seller and carrier.  The notice of the seller to the buyer that the product
has been shipped, creates an expectation in the buyer that he can query the
carrier regarding the transportation status.
 
Fred

-----Original Message-----
From: Nickolas Kavantzas [mailto:nickolas.kavantzas@oracle.com]
Sent: Tuesday, July 01, 2003 4:38 PM
To: jdart@tibco.com
Cc: Martin Chapman; Steve Ross-Talbot; Yaron Y. Goland;
public-ws-chor@w3.org
Subject: Re: Revised: Mission Statement


Well said Jon. 
  

To re-iterate a global model describes the common behaviour of a set of
interacting participants, so we may also call it a common or neutral 
view model. 


The message sends + receives of the individual participants and the wiring
between them are NOT modeled in the global model. 


What is modeled is the common, observable message interactions between
participants. 
  


Jon Dart wrote: 


Several statements have been made about the kind of model we don't want. 
But IMO it is not really clear enough what we do want. 

If I understand things correctly (a fairly big "if"), one requirement is 
that there be basically one model for both client & server (or peer and 
its peer, if you want to be more egalitarian). This means that I don't 
need to model the messages one party sends and have a parallel model of 
what the other party is receiving. The choreography description I expose 
to my partners should be sufficient for them to interact with me. This 
doesn't imply that there's one big model of all participants' message 
flows - in fact I think you don't want this. But it does imply that as 
party A directly interacting with party B, both parties have a model 
they can both view and base their interactions on (could include > 2 
participants also). It is a global model only in the sense that it 
encompasses directly interacting partipants, not the whole closure of 
participants with which those in turn interact (those are in separate, 
but related choreographies). 


--Jon 


Martin Chapman wrote: 
> I thought by our discussion at the F2F that the "in the middle of all of 
> them" is not what we are after, and in effect equates to the O word. So I 
> would be interested in a better understanding of what composition means, 
> given that this approach would not really support a wrapper wsdl. I'm not 
> arguing against composition here, just asking for some clarity. 
> 
> Martin. 
> 
> 
>>-----Original Message----- 
>>From: public-ws-chor-request@w3.org 
>>[ mailto:public-ws-chor-request@w3.org
<mailto:public-ws-chor-request@w3.org> ]On Behalf Of Steve Ross-Talbot 
>>Sent: Tuesday, July 01, 2003 10:52 AM 
>>To: Yaron Y. Goland 
>>Cc: public-ws-chor@w3.org 
>>Subject: Re: Revised: Mission Statement 
>> 
>> 
>> 
>>+1 
>> 
>>Having some understanding of the achitecture of this thing that we are 
>>doing 
>>would help us stay on the same page. 
>> 
>>If we have a global model then where does it sit with respect to any N 
>>participants. 
>>Is it in the middle of all of them? 
>>Is it at each one's site? 
>>Is it a proxy sort of thing? 
>>Is it a web service itself? 
>> 
>>What happens when we have two of them and we wish to compose? 
>>What happens to the originals? 
>>Where does the new one reside? 
>> 
>>I could go on. It would just be nice to get a common understanding. I 
>>have mine 
>>but I'm not sure it's the same as other peoples. 
>> 
>>Cheers 
>> 
>>Steve T 
>> 
>>On Tuesday, July 1, 2003, at 06:33  pm, Yaron Y. Goland wrote: 
>> 
>> 
>>>The key issue for me is what does it mean to compose a web service? 
>>>Does 
>>>this mean a new WSDL with some computer behind it that then forwards 
>>>requests to existing web services? Does this mean that a client is 
>>>expected 
>>>to send messages to different WS who all have some kind of 
>>>relationship with 
>>>each other? It's so vague that I'm not sure what scope we would be 
>>>signing 
>>>up for. 
>>> 
>>> 
>>> 
>>> 
>>>>-----Original Message----- 
>>>>From: Monica J. Martin [ mailto:monica.martin@sun.com
<mailto:monica.martin@sun.com> ] 
>>>>Sent: Monday, June 30, 2003 2:22 PM 
>>>>To: Yaron Y. Goland 
>>>>Cc: Francis McCabe; Burdett, David; Bonneau, Richard; Assaf Arkin; 
>>>>Jean-Jacques Dubray; public-ws-chor@w3.org 
>>>>Subject: Re: Revised: Mission Statement 
>>>> 
>>>> 
>>>> 
>>>>> 
>>>>>Goland: I don't understand what the terms service composition 
>>>> 
>>>>and service semantics 
>>>> 
>>>>>mean. Could someone please define them? Monica provides a whole mess 
>>>>>of 
>>>>>definitions but having 10 definitions is just as bad as having none. 
>>>>> 
>>>>>mm1:  The definitions were a compilation on various types of 
>>>> 
>>>>composition from the team.  We have not settled on one 
>>>>definition, although I have provided one below that seems 
>>>>appropriate here for consideration.  The definitions provided 
>>>>span different areas of composition, and whether the team agrees 
>>>>they are all the same, I can not speculate on.  I think it 
>>>>evidences the multiple levels of discussions that are occurring. 
>>>>Don't shoot the messenger. I would propose: **A service 
>>>>composition is a composition of services that results in a new 
>>>>service. The new service can be the combination of distinct parts 
>>>>to form a whole of the same generic type. The web services could 
>>>>be combined to achieve a specific goal.*  *This integrates parts 
>>>>of the definitions of recursive, web service and choreography 
>>>>composition. 
>>>> 
>>>>> Monica 
>>>>> 
>>>> 
>>>> 
>>>This email is confidential and may be protected by legal privilege. If 
>>>you are not the intended recipient,  please do not copy or disclose 
>>>its content but  delete the email and contact the sender immediately. 
>>>Whilst we run antivirus software on all internet emails we are not 
>>>liable for any loss or damage. The recipient is advised to run their 
>>>own antivirus software. 
>>> 
>> 
>>This email is confidential and may be protected by legal 
>>privilege. If you are not the intended recipient,  please do not 
>>copy or disclose its content but  delete the email and contact 
>>the sender immediately. Whilst we run antivirus software on all 
>>internet emails we are not liable for any loss or damage. The 
>>recipient is advised to run their own antivirus software. 
>> 
>> 
> 
> 
> 
>
Received on Wednesday, 2 July 2003 16:12:21 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Saturday, 18 December 2010 01:00:24 GMT