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

Re: WS-CDL v1, last call doc (dec 8, 2004)

From: Gary Brown <gary@enigmatec.net>
Date: Fri, 10 Dec 2004 09:58:10 -0000
Message-ID: <004901c4de9e$c58f5300$4b00a8c0@LATTITUDEGary>
To: <public-ws-chor@w3.org>
Only comment is:

2.5.2.3 Interaction Syntax

·Within the send or the receive element(s) of an exchange element, the causeException attribute when set to "true", specifies that an Exception MAY be caused at the respective Roles. In this case, the informationType of the exchange element MUST be of Exception Type. The default for this attribute is "false"

Why was the change made to 'MAY' (from 'will') regarding an exception occurring - this makes it non-deterministic whether an exception will occur? This change was introduced between Nov 11 and Dec 3 drafts - probably relates to the same issue I raised previously regarding the exception behaviour around a record.

>From my perspective, if an 'exchange' occurs that has causeException set to true, this indicates that the resulting behaviour of this interaction being performed is to move the relevant roles into their Exception states. If this assumption is not true, then I think we will be confusing the choreography authors - and we should probably rename the attribute "mayCauseException".

In 2.5.4 (Assigning Variables) it clearly states that when the "causeException" is true that it moves into the exception state:

"·When the attribute causeException is set to "true" in a copy element, the Role specified by the attribute roleType gets into the Exception state after the assign activity has completed"

So there appears to also be a consistency issue regarding the semantics of 'causeException'.

Regards
Gary

  ----- Original Message ----- 
  From: Nickolas Kavantzas 
  To: member-ws-chor@w3.org 
  Sent: Wednesday, December 08, 2004 11:11 PM
  Subject: WS-CDL v1, last call doc (dec 8, 2004)


  WS-CDL v1, last call doc (dec 8, 2004):

  http://lists.w3.org/Archives/Public/www-archive/2004Dec/0019.html



  We would suggest to look into the .doc (the .pdf seems to have some truncating issues).



  Regards,

  Nick on behalf of the WS-CDL Editors
Received on Friday, 10 December 2004 09:58:27 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Saturday, 18 December 2010 01:01:15 GMT