W3C home > Mailing lists > Public > public-ws-chor@w3.org > November 2004

[ws-chor] 11/2/2004: Response (Proposal) for Issue 687.3

From: Monica J. Martin <Monica.Martin@Sun.COM>
Date: Tue, 02 Nov 2004 11:30:43 -0800
To: "WS Choreography (E-mail)" <public-ws-chor@w3.org>
Message-id: <4187E063.1060801@sun.com>

Reference: http://www.w3.org/Bugs/Public/show_bug.cgi?id=687
Item 1: closed
Item 2: closed
Item 3: see below

On 12 October, I sent a message to the public list re: Issue 687.3 and 
was asked to provide a proposal 26 October 2004. See below.

3. In the F2F in March 2004, we indicated we did not acknowledge 
'dependent' choreographies (the blue boxes) that exist in the package or 
root (blue box). How then can we handle dependencies between imported, 
performed and choreographies with root=false?

    3.1 The current 22 Sept 2004 specification indicates that the bind
    variables if available between the performing and performed
    should be of the same participant type. It is often the case that,
    for example, a retailer takes on several roles
    across these composed choreographies. The Seller for the sale, the
    distributor for the warehouse, etc.  Does the
    bind element enforce the same participant type via the semantics
    (The Role Types within a single bind element MUST be carried out by the
    same party, hence they MUST belong to the same Participant Type)?

     >>Proposal: Defer to discussion by Tony Fletcher proposal on
    Participant Type and Role Types. I believe my example above is
    relevant to his proposal. Reference:

    3.2 In the Choreography package, if we do not have priority, can we
    enforce that the performed choreography can not be used unless
    enclosed in the enclosing choreography? If the answer is yes and the
    semantics support that, we can close 687.3.  Otherwise, a priority
    or an enforceable sequence may be required.  I am attempting to
    differentiate two choreographies in progress in the same package vs.
    one choreography in use and completes before the next choreography
    does. If we answer, we can also close Issue 615. Thank you.

     >>Proposal: Provide two examples in the specification like those
    included in reference that show how choreographies are performed.
    The two examples show sequential and parallel choreographies. This
    would result in no syntactic or semantic change (unless we consider
    strict sequence which has been a part of a recent discussion with
    Kohei Honda).

Received on Tuesday, 2 November 2004 19:30:43 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:01:06 UTC