Re: [ebxml-iic] Comments on ebMS Conformance Abstract Test Suite

["Michael Kass" <michael.kass@nist.gov>]

Mike,

  Thanks for reviewing the MS Conformance Test Suite Specification.
I've incorporated your comments and suggestions into the revised spec, which
I will post Sunday night.
  My responses are included below:

Thanks,
Mike

----- Original Message ----- 
From: "Mike Dillon - Drummond Group" <mike@drummondgroup.com>
To: <ebxml-iic@lists.oasis-open.org>
Sent: Monday, July 14, 2003 2:13 PM
Subject: [ebxml-iic] Comments on ebMS Conformance Abstract Test Suite


Hi all,

 I have walked through the suite and it looks very good !
 Very understandable.

 My detailed comments are very minor and are directly below.

 Thanks,
 Mike Dillon
 mike@drummondgroup.com
 817 875 0794

-------------------------------------------------------------------------
 Missing Tests
. SOAP Headers extensions with MustUnderstand of 0 may be ignored
There is not a test for this, probably not really important, but
  might be used a lot later for different "profiles" of ebMS

[MIKE] - This is a mising requirement and test, and I added it to
Conformance Test Requirements,
Abstract and Executable Test Suite.


 Mistakes
. Test 33 says that PartyID Type must be a URI, spec says it only MUST be a
string

[MIKE] - Test requirement specifies this is "recommended" it be a URI..
abstract and executable tests will
verify if it is or is not a URI, and will pass or fail the implementation
for this this "recommended" requirement

. There are 2 test #57s
. There are 2 test #67s

[MIKE] - Fixed

 Questions
. Not clear what test #109 is saying, If A sends B duplicate elimination
request,
that causes an element to appear in ack ?.

[MIKE] - Good point. This gets to the heart of the issue regarding what is
an "application level" requirement,
and what is an MSH requirement.   This test, as written assumes that
configuring an MSH with a CPA in which
DuplicatElimination is set to "always" would require an MSH to make sure
that any messages it sends contains a
DuplicateElimination element.  But if the CPA is nothing more than a
contract that is enforced at the application level,
then we can  not write a conformance test against the candidate MSH that can
test this, because inclusion of a
DuplicateElimination element is not controlled by the MSH.  I believe that
this issues is up for discussion on the conference
call on Monday.    So I will defer removing this test until it is decided
whether we can test any requirements that
are really at the "application level".


You may leave a Technical Committee at any time by visiting
http://www.oasis-open.org/apps/org/workgroup/ebxml-iic/members/leave_workgroup.php