Minutes
Oracle, Redwood Shores California
Opening
Agenda review
Jacques Durand suggested that we look at what WS-I has done related to the BP profile and automated testing
This will be inserted after morning break in today's meeting
Agenda agreed with above amendment
Resolution: m:Edwards s:Rutt minutes of 2008-01-15 approved w/o
Action: id=2008-01-15-1 status=done
Conformance Statements and Test Suites
Chapman:
we must go through the spec to use normative language
... the rule is very broad and we need to inform ourselves with more specific guidance.
Edwards:
We ought to find a way to connect each conformance test with a conformance statement contained and readily identifiable within
the spec
... perhaps a book/chapter/verse style of notation may be convenient.
Rutt:
Notes that many definitions are essential and normative but may not include the words MUST or MAY
... likes the idea of marking of numbering the normative parts
Chapman displays examples of notations used by BPEL
<Bryan Aupperle>
Please tell us what section you are looking at.
<Tom Rutt>
Some normative requirements may only be tested by a customer using platform specific tools, which are not appropriate for
standardized testing (i.e., no standard artifacts or test insertion points that would map to "layers" or "internal interfaces"
found in all implementations of the specification)
Chapman:
Hierarchy 1) Is it testable 2) Can we define a test 3) do we want to define a test 4) how do we define a test
proposal motion - The specification editors look to the WS-BPEL specification as a style to implement in this specification
concerning labeling of normative text and conformance statements.
proposal motion - The specification editors look to the WS-BPEL specification as a style to implement in this specification
concerning labeling of normative text and conformance statements. Editors shall endeavor to number or otherwise identify these
labeled sections so that they would remain steady across versions of the specification.
amendment moved by Edwards, seconded by karmarker
Rowley:
I think that the amendment is premature
Resolution: amendment passes 15 to 5 with 2 absentions
<fred>
You can add another +1 -- I pressed the wrong button to drop mute & killed connection
Resolution: m:Rowley s:Kand The specification editors look to the WS-BPEL specification as a style to implement in this specification
concerning labeling of normative text and conformance statements. Editors shall endeavor to number or otherwise identify these
labeled sections so that they would remain steady across versions of the specification. w/o
meeting reconvenes after break
Jacques Durand on WS-I Testing and Conformance
<anish>
here is a link to an earlier version of BP 2.0:
<anish>
note that the link i pasted above is an earlier version of BP 2.0 that was made public, what Jacques has displayed in the
f2f room is a later version (which is member-only and therefore does not have a public URL)
Jacques presents the approach to testing taken by WS-I
folks comment that the in-line presentation of tests is ugly and an impediment to readability.
Jacques presents an example of the WS-I test log which Edwards described as an artifact container.
Edwards points out that this type of test might not be appropriate for this TC since there is no pile of xml available on
the wire for inspection.
Discussion concerned the possibility of generating standard instrumentation to support testability.
Action: Testing sub-committee organise themselves, select a chair, and report to the TC
Are there optional conformance statements?
Edwards:
Are there optional things and how are they expressed? For example, are there conformance profiles?
Discussion followed concerning interoperability and permissibility of optional features as it relates to runtime implementations.
Scribe notes that this sounds a lot like be strict in your generation and permissive in your consumption...
Test Suites
Action: Test team to get on with it
Meeting is called to order
Restructuring of the Specification
Editors review the re-structured document
Rowley:
Is it possible to make the proposed version a CD?
... so there is no base line CD at this time...
Karmarkar:
Do we want to add samples before we call it a CD?
Rowley:
I don't think we should hold off until it is done
Edwards:
But it is not useful with chunks of normative stuff missing. We are still downgrade from the OSOA version.
Resolution: m:Freund s:Edwards Editor's proposal for document re-structuring is accepted w/o
Dynamic Aspects ...
Edwards:
the dynamic aspects of the domain means that things can come and go, but there is de minimis text describing such
Edwards steps up to the flip chart, armed with markers, and offers to collect requirements
<Mike Edwards>
Need good terminology
Requirement to be able to change aspects of existing deployed elements
Need to describe the process of:
Adding elements to the domain
Removing elements from the domain
Replacing an element with a different version
Need mechanics to allow reporting on the status of deployment changes (eg error reporting)
Bob supposes that for each contribution there exists a composite such that the integral of the composite from t=0 to t=infinity
approaches zero as the size of the contribution approaches zero
Discussion follows about the degree to which dynamic change should be discussed, allowed, or the mechanism specified.
Action: Edwards to write up one or more issues related to this topic.
Rowley:
Address what happens to autowire with newly deployed components and what happens to newly deployed wires.
bullet-How much to make normative vs allowing flexibility
Issue Assembly-3
Assembly-5
<Simon Nash>
NCName ::= (Letter | '_') (NCNameChar)*
Resolution: m:Freund s:Scott motion to defer the motion to close with no action w/o
Assembly-8
This one of Henning's issues and he is on leave for six months
Action: Rowley to propose a resolution to Assembly-8 according to his inclinations
Assembly-9
Resolution: m:Karmarkar s:Patil accept proposal to resolve Assembly-9 at http://lists.oasis-open.org/archives/sca-assembly/200801/msg00033.html
assuming that references in wd02.pdf are used w/o
Assembly-10
Action: Edwards to refine proposal to Assembly-10 and cross-reference to issue 6
Assembly-12 previously closed in 2007-11-06-3
Assembly-13 Difficulty using source attribute to pull values into complex types
Resolution: m:Chapman s:Malhotra Close with no action w/o
Assembly-14 Conflicting Specification of Values for Many-Valued StringProperties
Action: Karmarkar to coordinate with Alex Yu and produce a new proposal
Assembly-15 Unclear how Composites are discovered when Referenced
Resolution: m:Rowley s:Beisiegel Close Assembly-15 as a duplicate of Assembly-8 w/o
Assembly-9
Action: Beisiegel to raise an issue to fix those examples lacking required prefixes
Assembly-16 Component URI is not well described
Action: Rowley to propose a resolution to Assembly-16
Assembly-17 Need to define Namespace handling for included Composites
Action: Karmarkar to review and refine proposal to Assembly-17
Assembly-18 XSD definitions of Component Service and Component Referencehave unintended features
Action: Edwards to prepare a proposal using xsd to resolve Assembly-18
Assembly-19 Schema definition for Component, Component Reference @autowireattribute is incorrect
<anish>
<attribute name="autowire" type="boolean" use="optional" default="false"/>
<anish>
<attribute name="autowire" type="boolean" use="optional"/>
Resolution: m:Rowley s:Karmarkar resolve assembly-19 by removing the default on the autowire xml schema definition w/o
Assembly-20 Conflict Between SCA XSD and Specification Body about the handling of Multi-Valued Properties
Resolution: m:Karmarkar s:Rowley Close Assembly-20 as a duplicate of Assembly-14 w/o
Assembly-21 Binding name attribute not a QName
Resolution: m:Rowley s:Edwards Change pseudo schema for the name attribute of binding to be an NCname w/o
Assembly-22 WSDL extension should not be required for conversations
Action: Booz to generate a proposal for the resolution to Assembly-22
Review of tomorrow's agenda
meeting in recess until tomorrow.
Continuation of meeting of 2008-01-23
Callbacks and conversational Interfaces because the BPEL people are causing trouble
Edwards:
Do we have it right or is it too complex?
... can it be implemented using common communications methods?
Discussion continued with members expressing hope for interoperability but with a sense that total interoperability would
be an unrealistic goal and limited interoperability might still be successful
Other concerns are that involvement by application layers make conversational interface interoperability more difficult unless
all of the functionality were delegated to those application layers. Dividing the responsibility is rife with hazards.
Edwards:
Is it confusing when callbacks & conversational interfaces are mixed together? Should they be mutually exclusive?
Koenig presents his slides illustrating some of the issues.
Action: Koenig to produce a proposal to address Assembly-33 based in part on his presentation and the discussion in the room
Edwards:
Is i posible to implement callbacks and/or conversational interfaces uning common communications methods?
... solution-- provide exemplary implementations for each of them in Web services, JMS
Action: Edwards to take this topic to the bindings tcs for further consideration
Assembly-4 Dynamic Change of Callbacks
Peshev:
describes his proposal and rationale thereof
Motion: m:Rowley s:Nash Delete the last paragraph of WS-02 8.2 numbered ca lines 2241-2244
Amendment: m:Edwards s:Rowley Replace the word implement with the word provides on line line Nr. 2237
Amendment: withdrawn
Amendment: m:Rowley s:Edwards In line 2237 replace the word implement with "provide an implementation of"
Amendment: passes 8 to 4 with 8 abstentions
Amendment: m:Nash s:Rutt replace provide an implementation of with provide
Amendment: fails 1 to 6 with 11 abstentions
Amendment: m:karmarker s:Booz in addtion to the motion previously stated, additionally In line 2237 delete the phrase "component implementation"
<Sanjay>
pleasures of wordsmithing by committee
Amendment: passes 11 to 1 with 2 abstentions
Motion: restated as amended -- Delete the last paragraph of WS-02 8.2 numbered ca lines 2241-2244 and in addition line 2237 replace
the word implement with "provide an implementation of" and in addition In line 2237 delete the phrase "component implementation"
Amendment: m:Rowley s:Edwards resolve Assembly-4 by...
Amendment: passes w/o
Resolution: Resolve Assembly-4 with motion above as amended w/o
Issue 1 solicitation of concerns
Break for lunch until 13:15 Pacific
Restarting from the break
Issue 1
Motion: m:Rowley s:Vorthmann we disallow bindings from references unless the binding specifies a URL or some other way of specifying
the endpoint
<Michael Rowley>
If a reference does not have a binding, then the binding used can be any of the bindings specified by the service provider,
or other means, as long as the intents required by the reference and the service are all respected.
Amendment: m:Sanjay s:Rowley If a reference does not have a binding, then the binding used can be any of the bindings specified by the
service provider, or binding.sca, as long as the intents required by the reference and the service are all respected.
Amendment: fails 7-7-5
Amendment: m:Nash s:Edwards If a reference does not have a binding, then the binding used can be any of the bindings specified by the
service provider, as long as the intents required by the reference and the service are all respected.
Amendment: passes 11-2-3
Amendment: m:Karmarkar s:Nash remove the text "we disallow bindings from references unless the binding specifies a URL or some other
way of specifying the endpoint" from the amended motion
if this amendment passes the amended motion would be:
"If a reference does not have a binding, then the binding used can be any of the bindings specified by the service provider,
as long as the intents required by the reference and the service are all respected."
the current amended motion is:
"If a reference does not have a binding, then the binding used can be any of the bindings specified by the service provider,
as long as the intents required by the reference and the service are all respected."
Motion: If a reference does not have a binding, then the binding used can be any of the bindings specified by the service provider,
as long as the intents required by the reference and the service are all respected.
Resolution: motion passes as amended 14-3-2
<Martin C>
please ensure your name has been recorded!
Motion: m: Chapman s: Rowley resolve issue 1.A with the previous resolution/motion and issue 1.B will be considered duplicate of 26
Resolution: motion passes 11-2
issue 31
Motion: m:Rowley s: Rutt resolution to issue 1 already resolves issue 31. So resolve issue 31 by pointing to resolution of issue 1
Motion: m: Rowley s: Rutt to defer the current motion on the table
Motion: to defer fails 2-5-1
Motion: m:Nash s:Rutt calls the question
Motion: calling the question fails 4-4
Resolution: motion fails 6-6-1
issue 24
issue 24 defered till anish finishes his AI
issue 27
Motion: m:Nash s: Beisiegel delete lines 2537-2541 in WD-02.pdf and resolve issue 27
Resolution: delete lines 2537-2541 in WD-02.pdf and resolve issue 24 w/o
1) Fold in resolved issues
suggestion to do milestone (3) for a section and then get approval from the TC. And move on to the rest of the document and
tackle it in manageable chunks
{milestone 3: 3 weeks from today}
7.a) plan for testing in 2 weeks
testing subcommittee: tentative call date/time -- monday 8am or wednesday 8am pacific
Motion: m: Karmarkar s: Edwards motion to adjourn
Resolution: to adjourn w/o
Schreiber diagnostics output
[Delete this section before publishing the minutes]
statistics: Schreiber found 257 input lines
edits: Schreiber found the following text-edit commands:
edits: Line 40: Bob: s/break/morning break
edits: Line 53: Bob: s/ot/of
edits: Line 63: Bob: s/specificatoin/specification
edits: Line 64: Bob: s/pamendment/amendment
edits: Line 71: Bob: s/Testing/WS-I Testing
edits: Line 108: Bob: s/to r/to allow r
edits: Line 126: Bob: s/10/10 and cross-reference to issue 6
edits: Line 129: Bob: s/13/13 Difficulty using source attribute to pull values into complex types
edits: Line 132: Bob: s/11-06/11-06-3
edits: Line 138: Bob: s/Rowley/Beisiegel
edits: Line 143: Bob: s/-7/-17
edits: Line 154: Bob: s/solution/resolution:
edits: Line 155: anish: s/Karmarker/Karmarkar/G
edits: Line 159: Bob: s/autowirexml/autowire xml
edits: Line 171: Bob: s/conversation/conversational
edits: Line 179: Bob: s/i right/it right
edits: Line 213: Bob: s/provides/provide
edits: Line 217: Bob: s/absentions/abstentions
edits: Line 221: Bob: s/implements/provide an implementation of
edits: Line 308: anish: s/defer/defered/
edits: Line 314: anish: s/issue 24/issue 27/
edits: Line 338: anish: s/4/5/
command-scribe: Line 34: Bob Freund recognized
command-scribe: Schreiber detected that this section was scribed online
edit-substitute: command on line 40 succeeded, changed line 39 from 'break' to 'morning break'
edit-delete: Line 40 was deleted
edit-substitute: command on line 53 succeeded, changed line 52 from 'ot' to 'of'
edit-delete: Line 53 was deleted
edit-substitute: command on line 63 succeeded, changed line 61 from 'specificatoin' to 'specification'
edit-substitute: command on line 64 succeeded, changed line 62 from 'pamendment' to 'amendment'
edit-delete: Line 63 was deleted
edit-delete: Line 64 was deleted
edit-substitute: command on line 71 succeeded, changed line 70 from 'Testing' to 'WS-I Testing'
edit-delete: Line 71 was deleted
edit-substitute: command on line 108 succeeded, changed line 107 from 'to r' to 'to allow r'
edit-delete: Line 108 was deleted
edit-substitute: command on line 126 succeeded, changed line 125 from '10' to '10 and cross-reference to issue 6'
edit-delete: Line 126 was deleted
edit-substitute: command on line 132 succeeded, changed line 127 from '11-06' to '11-06-3'
edit-substitute: command on line 129 succeeded, changed line 128 from '13' to '13 Difficulty using source attribute to pull
values into complex types'
edit-delete: Line 129 was deleted
edit-delete: Line 132 was deleted
edit-substitute: command on line 155 succeeded, changed line 133 from 'Karmarker' to 'Karmarkar'
edit-substitute: command on line 138 succeeded, changed line 137 from 'Rowley' to 'Beisiegel'
edit-delete: Line 138 was deleted
edit-substitute: command on line 143 succeeded, changed line 141 from '-7' to '-17'
edit-delete: Line 143 was deleted
edit-substitute: command on line 159 succeeded, changed line 151 from 'autowirexml' to 'autowire xml'
edit-substitute: command on line 154 succeeded, changed line 153 from 'solution' to 'resolution:'
edit-substitute: command on line 155 succeeded, changed line 153 from 'Karmarker' to 'Karmarkar'
edit-delete: Line 154 was deleted
edit-delete: Line 155 was deleted
edit-delete: Line 159 was deleted
command-scribe: Line 166: Bob Freund recognized
command-scribe: Schreiber detected that this section was scribed online
edit-substitute: command on line 171 succeeded, changed line 169 from 'conversation' to 'conversational'
edit-delete: Line 171 was deleted
edit-substitute: command on line 179 succeeded, changed line 175 from 'i right' to 'it right'
edit-delete: Line 179 was deleted
edit-substitute: command on line 213 succeeded, changed line 211 from 'provides' to 'provide'
edit-delete: Line 213 was deleted
edit-substitute: command on line 217 succeeded, changed line 215 from 'absentions' to 'abstentions'
edit-delete: Line 217 was deleted
edit-substitute: command on line 221 succeeded, changed line 219 from 'implements' to 'provide an implementation of'
edit-delete: Line 221 was deleted
command-scribe: Line 243: Anish Karmarkar recognized
command-scribe: Schreiber detected that this section was scribed online
edit-substitute: command on line 308 succeeded, changed line 306 from 'defer' to 'defered'
edit-delete: Line 308 was deleted
edit-substitute: command on line 314 succeeded, changed line 312 from 'issue 24' to 'issue 27'
edit-delete: Line 314 was deleted
edit-substitute: command on line 338 succeeded, changed line 336 from '4' to '5'
edit-delete: Line 338 was deleted
system: Transformer: SAXON SA 8.9
[End of Schreiber diagnostic output]