[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Raw minutes for today
Simon Holdsworth: Participant Code: 7059536 USA Toll-Free 888-426-6840 USA Caller Paid 215-861-6239 UNITED KINGDOM Toll-Free 0800-368-0638 UNITED KINGDOM Caller Paid 0-20-30596451 Ireland Toll-Free 1-800-943-427 Ireland Caller Paid 0-1-5264424 Bulgaria Toll-Free 00800-117-4514 Other access codes can be found at: https://www.teleconference.att.com/servlet/glbAccess?process=1&accessCode=7059536&accessNumber=02030596451
Simon Holdsworth: Agenda
Simon Holdsworth: 1. Opening Introductions Scribe assignment Top of the scribe list: Plamen Pavlov SAP AG David Booz IBM Tom Rutt Fujitsu Limited Ashok Malhotra Oracle Corporation Bryan Aupperle IBM Eric Johnson TIBCO Software Inc. Anish Karmarkar Oracle Corporation Martin Chapman Oracle Corporation Agenda bashing 2. Approval of the minutes from 26 May: http://www.oasis-open.org/committees/download.php/42365/SCA%20Bindings%20minutes%202011-05-26.doc 3. Actions 20110519-01 [Simon Holdsworth] Look into whether we need a testcase for the updated normative statement in BJM60009 [following resolution of BINDINGS-157]. 4. New Issues No new issues 5. Open Issues http://www.osoa.org/jira/browse/BINDINGS-153 write new tests for TA-20021 TA-20022 TA-20028 TA-20032TA-20034 TA-20035 TA-40007 TA-50008 TA-50009 Status: Proposed outline resolution in JIRA; partial direction agreed for some statements. 6. AOB
Please change your name from 'anonymous' using the Settings button
[12:04] Bryan Aupperle: Scribe: Bryan Aupperle
[12:04] Bryan Aupperle: Meeting is quorate (66%)
[12:05] Bryan Aupperle: Topic: Agenda bashing
[12:05] Bryan Aupperle: Agenda agreed
[12:05] Bryan Aupperle: Topic: Approval of Minutes
[12:06] Bryan Aupperle: Resolution: Minutes approved w/o
[12:06] Bryan Aupperle: Topic: Action Items
[12:06] Bryan Aupperle: 20110519-01 - Pending. To be discussed later in call
[12:07] Bryan Aupperle: Topic: Bindings 153
[12:07] Bryan Aupperle: http://www.osoa.org/jira/browse/BINDINGS-153 write new tests for TA-20021 TA-20022 TA-20028 TA-20032TA-20034 TA-20035 TA-40007 TA-50008 TA-50009 Status: Proposed outline resolution in JIRA; partial direction agreed for some statements.
[12:08] Bryan Aupperle: Mike: I have sent a note to the list of the optional normative statements in the WS spec.
[12:09] Bryan Aupperle: Simon: There is significant overlap in this list with the normative statements in issue 153.
[12:09] Tom Rutt (Fujitsu): http://www.oasis-open.org/apps/org/workgroup/sca-bindings/email/archives/201106/msg00002.html
[12:10] Bryan Aupperle: Mike: I did not realize that we had discussed some of these statements already.
[12:10] Bryan Aupperle: Mike: The issue not really not about test cases, more about the optional statements.
[12:12] anish: i disagree, the core issue is about tests and not about optionality. They are related, but different.
[12:12] Bryan Aupperle: Mike: I would prefer to have a separate issue targeted at the spec.
[12:13] Bryan Aupperle: Simon: There are a few differences. And neither is a subset of the other.
[12:14] anish: for some of the stmts in issue 153, we converted should/mays to musts and some we removed them
[12:14] Bryan Aupperle: Mike: Did we cover the complete list in issue 153?
[12:14] anish: i.e., in terms of direction, that is
[12:14] Bryan Aupperle: Simon: We did not.
[12:15] Mike Edwards: Mike's email about optional normative statements: http://lists.oasis-open.org/archives/sca-bindings/201106/msg00002.html
[12:16] Bryan Aupperle: Anish: We should finish going through 153.
[12:17] Bryan Aupperle: Anish summarizes the outcome of the discussion last week.
[12:20] Mike Edwards: My initial proposal is to make BWS40007 non-normative
[12:20] Bryan Aupperle: Discussion resumes on BWS40007...
[12:22] Bryan Aupperle: Anish: We should convert this to a MUST for interop.
[12:23] Bryan Aupperle: Simon: We should have Eric's input and he is not on the call.
[12:24] Bryan Aupperle: No conclusion reached.
[12:24] Bryan Aupperle: BWS50010
[12:27] Bryan Aupperle: Anish: This is a MUST statement. Should add a test using the WS policy assertion.
[12:29] Bryan Aupperle: Mike: For this to be testable there would have to be an alternative callback protocol that could be used in an observable manner.
[12:30] Bryan Aupperle: Mike: For example, Tuscany always used the protocol for callback independent of the policy assertion.
[12:34] Bryan Aupperle: Mike: And actually this brings up a bigger concern. WSCB itself is optional, and we not have a defined way to differentiate the different failure modes in the test suite.
[12:34] Bryan Aupperle: Anish: There needs to be a list of WSCB tests that only need to be run if an implementation claims to support WSCB.
[12:36] Bryan Aupperle: Mike: Even with that, it is not clear how to write a test detecting which protocol might be used if an alternative protocol were to be developed.
[12:38] Bryan Aupperle: Mike: It is not untestable, just very difficult to test.
[12:41] Bryan Aupperle: Anish: Adding an assertion to a WSDL should be done. But I see that this is not a complete test.
[12:41] Bryan Aupperle: More consideration needed.
[12:42] Bryan Aupperle: BWS50013
[12:42] Mike Edwards: "SCA runtimes are not required to support or understand WS Policy - untestable"
[12:42] Bryan Aupperle: Anish: This could be covered by a simple static test.
[12:43] Bryan Aupperle: Mike: WS-Policy support is not required so we cannot write a meaningful test.
[12:44] Bryan Aupperle: Ashok: Correct. SCA intents have to be supported but we do not require any particular policy language.
[12:46] Bryan Aupperle: Anish: So the question is if a runtime supports WSCB, is it required to support WS-Policy and the assertion?
[12:46] Bryan Aupperle: Anish: This is not covered in the normative statements, but I need to review the explanatory text.
[12:48] Bryan Aupperle: Mike: This gets back to the overall optionality of WSCB and the need to understand what failing a test means.
[12:48] Bryan Aupperle: Anish: I do not find any requirement for supporting WS-Policy if WSCB is supported.
[12:50] Bryan Aupperle: Anish: Key is taht runtime must not silently ignore the policy assertion.
[12:52] Bryan Aupperle: Ashok: As I understand the problem, we do not require support for WS-Policy but we use WS-Policy to control the use of WSCB. So an implementation that uses a different policy language need not understand the policy assertion.
[12:53] Bryan Aupperle: Simon: This will take more discussion.
[12:53] Bryan Aupperle: Ashok: It would help if someone created a straw man to start the discussion.
[12:54] Bryan Aupperle: Tom: We could require understanding just the one assertion without full WS-Policy support.
[12:54] Bryan Aupperle: Mike: Again, we do not require use of or even support of the assertion.
[12:56] Bryan Aupperle: Anish: Problem with requiring support for just the one assertion it that it is contained within a s PS-Policy element. You really have to support WS-Policy to be able to even understand the assertion.
[12:57] Bryan Aupperle: Anish: WSDL requires can be a help here. I.e. if you do not understand WS-Policy and WSCB then you must not accept the WSDL.
[12:58] Bryan Aupperle: Simon: We need to continue this discussion.
[12:58] Bryan Aupperle: Topic: AOB
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]