Minutes W3C WS Choreography WG con call held on 10th June 2003, 1pm PDT.

Agenda:


1.            Role Call

2.            Confirm scribe

3.            Approve minute

        3rd June minutes: http://lists.w3.org/Archives/Public/public-ws-chor/2003Jun/0012.html

 

4.             Action Item Review

 

 

 

5.             Tracking items (a section designed to ensure that longer running items are properly tracked and issues collected)


·                     Mission Statement (progress)

·                     Requirements (progress)

·                     WS-BPEL co-chairs at our face to face (working out the basis for understanding)

6.             Decision Points and partial execution (from email discussion) – continued from last week

·                      Now has become Combining Policies

7.            Threads (on the archive not covered elsewhere in the agenda)

1.                   Correlation (Why and How?) ….

2.                   Requirements from JJD/Asaf/…. 

3.                   Is choreography a contract?

8.            AOB.

Role Call:

 

Chairs:

 

 Martin Chapman

Oracle

Steve Ross-Talbot

Enigmatec

 

 

W3C Staff Contacts

 

Yves Lafon

 

 

 

 

 

 

Members:

 

Yaron Goland

BEA Systems

Mayilraj Krishnan

Cisco Systems Inc

David Burdett

Commerce One

Francis McCabe

Fujitsu Ltd

Yoko Seki

Hitachi, Ltd.

Ravi Byakod

Intalio Inc.

Eunju Kim

National Computerization Agency

Greg Ritzinger

Novell

Nickolas Kavantzas

Oracle Corporation

Jeff Mischkinsky

Oracle Corporation

Ugo Corda

SeeBeyond Technology Corporation

Michael Champion

Software AG

Monica Martin

Sun Microsystems, Inc.

Jim Hendler

University of Maryland (Mind Lab)

Daniel Austin

W. W. Grainger, Inc.

Leonard Greski

W. W. Grainger, Inc.

Ed Peters

webMethods, Inc.

 

 

Raw IRC log at:  http://www.w3.org/2003/06/10-ws-chor-irc

 

Appointment of scribe:

 

Jean-Jacques Dubray, Eigner, kindly volunteered to scribe for the meeting.

 

Previous scribes (in order):  Monica Martin, Mayilraj Krishnan, Francis McCabe, Michael Champion, Abbie Barbir, David Burdett, Jon Dart, Carol McDonald, Yaron Goland, Leonard Greski, Ed Peters, Greg Ritzinger, Daniel Austin, Peter Furniss, Jim Hendler

 

 

Minutes Approval

Draft minutes at  http://lists.w3.org/Archives/Public/public-ws-chor/2003Jun/0012.html

 

Minutes approved, subject to adding Eunju Kim to the attendance list.

 

Action Item Review

 

ACTION:   Carol to look into WSCI summary.

Carol is not here, no progress

ACTION:   SRT to capture reqs from minutes.

No progress

ACTION:   All: alternate formal models other than the pi-calculus. Agenda item for next F2F – proposal from FM  Event Calculus)

Ongoing. Other formal models other than PI: only one so far: Event Calculus

§         Would be good to have an email for other proposals

§         Frank to present event calculus at F2F

ACTION:   Yves to setup reading list page. SRT will work with Yves.

Done: Reading list page is now set up

ACTION:   DA  will do the wordsmith, summarize and put together revised mission statement. Once again it is not final.

Ongoing. Daniel: wordsmithing around the mission statement, will send it out by the end of the day.

ACTION:   SRT to extract summaries of harvested specs (wsci, bpel).

Ongoing:WSCI, BEPL: harvesting these specs: (SRT) what is the focus of the action items?

o       Use cases,

o       State of the art review (MC)

o       So the action item is to collect the summary from other people about these specs

ACTION:  Co-chairs to harvest agenda items for F2F

   Done

ACTION: DA will own the privacy discussion thread and summarise the requirements that ensue. http://lists.w3.org/Archives/Public/public-ws-chor/2003May/0167.html

 

Ongoing. Daniel doesn’t really own privacy but will extract requirement into requirements doc.

 

FMC -> look at the Mike Mehan privacy doc for the WSA document

New ACTION: FMC to send in URL for WSA privacy document

 

ACTION:  Grainger to ask for repost on F2F logistics.

 

·        Reposting of the f2f logistic (Done)

o       There will be an update later with a greater level of detail.

 

 

 

 

F2F: proposed agenda for next week:

o       Day 1 is mostly requirements

§         It is really difficult to track subjects when we do not pay attention to the subject.

§         This issue will be covered in the agenda

§         Go over the draft document from Daniel

§         Requirements for other working groups (WSDL, WSA, …)

o       Day 2

§         Review the issues

§         BPEL liaison (chairs are coming)

§         Maybe have 30 minutes of preparation for this summit

§         Formal model: get some agreement, how formalism would affect the language. Will add Frank’s presentation on Event Calculus to the Agenda (30 minutes would be enough).

§          

New ACTION: MC to add FMC to F2F agenda for event calculus presentation

 

§         What features we need in the language (bottom up)

·        David

·        JJ (BPSS over again, what would be done differently)

 

-         WS-BEPL co-chairs being there, it was suggested that we think how to present ourselves to them

o       A single page could be prepared to express what we are doing

o       (MC) talk about it on the first day of the f2f

o       (SRT) takes the action to write this page (draft). Targeting having  liaison with BPEL. Draw a line for this liaison.

 

New ACTION: SRT to write a straw man proposal on BPEL liaison.

 

 

§         sent it to the mailing list

§         should be the public list with careful wordings

§         the member list is closed to the press        but still open to most people

§         So this is will be to send to the members list

 

Requirements

 

Where are we? Understand what you can send out and when ahead of f2f. DA: we will have something by Friday.

Abbiw has done his part and has collected most of the use cases.

It won’t be complete, but food for thoughts at the meeting.

 

 

Decision points and partial execution

o       It moved into the topic on combining policies

o       DB: what do you make visible in a choreography

§         Don’t have to make public how the decision was made

§         It moved into combined policies because we want to use information about the choreography they support in order to do business with them. Hence you may need to know how they do security, … i.e. policies, and therefore, what combination of policies that need to be in place to be able to do business

§         DA: make a series of assertions about the choreography, is it part of our scope? DB, not at all, MC: should be part of our features discussion.

§         What are the dependencies with other specifications? Can we rely on them.

§         DA/MC: cannot reference anything that is proprietary or not done, or not sure it will be done.

§         DA: do we bounce this back to WSA such that new working group can be efficiently spawned.

§         Frank: we need to make a case for the WSA (e.g. policy and contract working group). We need to write a charter for such a group in order to speed up the decision

§         Mike: It was talked in WSA, but there are too many things going on, there is a lot of politics, who wants to own it. Another remark is that the scope of this group is quite open, so we could bring it in scope

§         DB: disagree to bring in scope.

§         MC: we establish the requirements and then we decide which group will do it. Talk about this on the F2F agenda.

 

Correlation

o       Why is it important?

o       How do we approach it?

o       Yaron:

§         Uber generic mechanism: set of XPath that look at an incoming message and determine the correlation of this message so that it can be related to a choreography instance.

·        The problem, it is hard to program, lots of record keeping,

·        Most people screwed up

§         SOAP based mechanism, then the correlation is done on this id.

·        Normal people can use this mechanism

§         DB: Correlation is 1) identify a message so that it can be referenced later, 2) correlate two messages together (request/response), 3) long running conversation

·        That’s different from looking inside the message, which has to do with routing rather than identifying if a message is related to a choreography instance

·        At C1, do it with SOAP, which choreography definition is being used, then a conversation id (choreography instance id).

·        JJ: could also need the MEP id.

·        SRT: what are the impact / requirements on this group?

·        Yaron: at the choreography we only need to assert that there exist the mechanism by which the correlation is interpreted.

·        DB: we should have enough abstraction so that it is not tied to specific technologies. If we define a binding to SOAP then we may want to specify how we do it.

·        SRT: hearing a number of requirements for relating messages to specific choreography instances. Need to distil the requirements, then we decide to how we go about solving requirements.

·        DB: this is a MUST HAVE for the spec

·        DA: even if it is out of scope, we need to define very well the interface between our spec and other group work.

 

Is choreography a contract

 

o       DA: it is not a contract, but some sort of contract needs to be in place

o       DB: a choreography definition is a specification

§         A contract is an agreement about interoperating with this definition

§         Hence, it is not a contract but part of a contract

o       MC: Think about WSDL as contract. They are technical contracts not legally binding contracts. “if I send you this request, you will respond like this…” this is a contract. This group’s scope is limited to technical contracts

o       Frank: Agree with Martin, there are enough similarities between legal and technical contracts that we could use similar concepts. A contract is not a piece of paper, but a social fact that the piece of paper gives evidence of. There is often confusion on the definition of a contract. The simplest way of thinking of it is to say that this is a social fact. A choreography could result in new agreement being established. We should be interested in contracts and related ideas. It is a deep well, if not a black hole

o       Jeff: computer programs, who does what, when, where, standard 101 design by contract.

o       JJ: it can have linkage to business contracts

o       Monica: it provides context on what is being executed (input, output, outcome).

o       The contract has to be enforceable.

o       MC: Can’t have the technical code reference a business contract.

o       DB: both approaches are relevant. Part of the contract can be a technical contract. CPA inside of ebXML. Will be put forward a definition of contract in eCommerce. Jeff could do the same for design.

o       MC: do we need legal stuff.

o       JJ: the state of the choreography must be unambiguous such that legal contract can be based on it.

 

AOB

 

-         No call next week. What about the week after. SRT: keep going.

-         Has anything been arranged for telecon at the f2f: there will be dial-in to the meeting (talk and hear). The Zakim bridge will be used. W3C won’t be able to provide it to the whole time. Yves will try to arrange that.

-         MC: will send an email to ask who plans on attending the dial-in

-         DA: Must register now.

-         SRT: more people will come UBS (Union Bank of Switzerland). Mr. (Tom?) Carol may come.

 

Summary of New Action Points

New ACTION: FMC to send in URL for WSA privacy document

New ACTION: MC to add FMC to F2F agenda for event calculus presentation

New ACTION: SRT to write a straw man proposal on BPEL liaison.