OASIS WS-TX - Minutes - February 23, 2006

 

Chat room: http://webconf.soaphub.org/conf/room/wstx

 

Highlights:

All TC members should vote on the ballot concerning attendance at the Face-to-Face (F2F) meeting March 14-15 in Raleigh, NC, USA.

http://www.oasis-open.org/apps/org/workgroup/ws-tx/ballot.php?id=915

 

New action Items:

ACTION: Ian and Eric to produce a list of guidelines or procedures for editors, by next Tuesday.

ACTION: Editors to produce updated Working Drafts by 7th

ACTION: Eric and Ian: create agenda for F2F by 7th.

ACTION: Eric and Ian to send a message confirming or canceling the March 9 teleconference meeting by the 6th.

ACTION: Chairs to inform ASIS that we are not following the guideline with respect to hyphens.

ACTION: Alastair, editors of AT and C and BA to develop concrete proposal to move material from section 9 of WS-AT on header elements to WS-Coordination and allow it be used by reference in WS-AT and WS-BA.

 

 

Agenda:

1. Roll Call - chair

Eric called the roll.  Initially 22/35 voting members, so the meeting is quorate.

 

Doug Davis: Eric - would it be possible for you to 'ack' each person as they say 'present'

- otherwise we don't know if you heard us?

 

2. Confirm Paul as minute taker - chair

Paul Knight taking minutes.

 

3. Approve minutes of Feb 09 telecon - chair

Alastair Green: In the minutes of the discussion of issue 007, Ian's response to my question about the request-reply MEP is missing. Please add:

 

"Ian: That is a very good question."

 

after:  Alastair: Why do we even have the request-reply MEP?

 

Eric: Any other changes for the minutes?  Can we adopt them as changed?

 

No objections.

Minutes adopted as changed.

 

4. Call for AOB - chair

none noted

  

5. Action Review - Paul

#0019: Editor to resolve inconsistencies from Issue 017. Owner: Max Feingold

Closed

#0017: Editor to resolve inconsistencies from Issue 017. Owner: Mark Little

Closed

#0014: Max and Alastair to develop text addressing issues 018, 019, 020, 021. Owner: Max Feingold

Material uploaded two weeks ago.  Closed.

 

6.  Potential new issues to accept - Colleen

Brief statement of each proposed new issue. There is no need to consider the merit of any proposed resolution at this stage. In general the TC will accept issues that are not reopening resolved issues and that are in-scope. Accepted issues move from "review" state to "active" state.

     i0xx  New issue....

http://docs.oasis-open.org/ws-tx/issues/WSTransactionIssues.xml

 

two new issues raised since the last call

Issue 24: http://www.oasis-open.org/archives/ws-tx/200602/msg00030.html

Ram Jeyaraman: (the issue author) The change is essentially editorial.

Accepted, moved to "active" state.

 

Issue 25:  http://www.oasis-open.org/archives/ws-tx/200602/msg00029.html

Minor editorial change.

Accepted, moved to "active" state.

 

 

7. Plan for first committee drafts

  Decide on when it makes sense to produce first committee drafts.

Eric:  We would very much like to get the committee drafts ready for the F2F meeting March 14-15, with the issues resolved as of today to be incorporated by March 6, in time for everyone to have a chance to review them.

 

Ian: What we are after is proposed, unapproved working drafts which we will attempt to approve as committee drafts at the meeting.

 

Eric: We need to vote on them in order to move them to CD status.

 

Ian: Yes, this will be our first set of committee-approved CDs.

 

Martin: We should focus on WS-Coordination before the transaction stuff.  We should get the CD done as early as possible.

 

Paul Cotton: Yes, as early and as often as possible.  Can the editors produce working drafts before March 7?  Then we will have a week for review and to prepare comments.  Is one week enough? 

 

Paul C: My second point is that actually voting something to CD requires a full majority of the TC, regardless of the number present at the meeting.

 

Ian: I don't believe that is correct.

 

Paul: I believe so, that is why the voting of many TCs is electronic, to engage the largest number possible. The editors will need to balance several things, finishing the CD, and tracking the changes.

 

Ian:  My apologies to Paul Cotton, he is correct and we do need 50% of the voting members.  I think we will have enough attendance to reach that level at the F2F.

 

*****************from chat**************

PaulC: Quoted from the OASIS TC Guidelines:

 

3.1 Approval of a Committee Draft 
 
The TC may at any stage during development of a specification approve the specification 
as a Committee Draft. The approval of a Committee Draft shall require a Full Majority Vote 
of the TC. The TC may approve a specification, revise it, and re-approve it any number of 
times as a Committee Draft. 

 

Martin Chapman: "Quorum" is the number of Voting Members of a TC that must be present in a 
meeting so that Resolutions and decisions may be made. The Quorum for OASIS TC meetings is a 
simple majority (more than half) of Voting Members. So if meeting is quorate and all vote 
yes (no one objects), that meets the full majority rule.

 

EricN: So assuming the F2F is quorate it would be possible to have a full majority vote.

 

Martin Chapman: Yes if there are no objections. If there are, it needs to be counted based 
on total voting list.

**************end from chat***************

 

Alastair: It may be premature to produce CDs for WS-AT and WS-BA.  We have been holding off submitting comments on them, since we have been focusing on WS-Coordination.  I agree with Paul that the F2F will continue WS-C issue resolution, and the WS-C CD should incorporate those changes.  A working draft includes all the editorial changes agreed by the TC.  We should encourage the editors to turn the handle to produce a CD for WS-C, but not necessarily for WS-AT and WS-BA.

 

Ian: If we are able to vote for the approval of any CDs at the F2F, we should do so.  I had been thinking of producing all three CDs, but we are indeed focusing on WS-C first.  Some of the issues, including the namespace URI issue, do affect all three specs.  It will be useful to produce CDs for all three, with the small number of changes we have agreed already.  From the perspective of reviewing the specs, Eric and I together will produce a set of guidelines for the editors to constrain the changes which we have discussed and resolved so far.  We will request the editors to use change bars to highlight the changes.

 

Colleen: Propose that we go into this with the goal of having each of the specs with a working draft, and have a checkpoint at the second day of the F2F to see if it makes sense to move the documents to CD shortly after the F2F.

 

Eric: That is a middle position between the other positions.  We have also heard some comments on whether it is reasonable to produce the CDs before the meeting.

 

Paul C: I reiterate: do it early and do it often (getting the documents out before the F2F).  Get the working drafts out, get CDs out.

 

Paul C: What would the diffs be against?  I suggest they should be against the first working drafts, where the editors applied the boilerplates.  The editors may want to produce multiple diff documents, to communicate the changes which have been applied.

 

Ian: The editors should all take the same tack on producing diffs.

 

Eric: Ian and I will produce some guidelines for the editors and reach agreement in the editor's subcommittee on what we will do.  We will ask the specification editors to produce updated working drafts by the 7th, using template and tracking changes.

 

ACTION: Ian and Eric to produce a list of guidelines or procedures for editors, by next Tuesday.

 

Eric: Will this time frame work for the editors?

Max Feingold: This sounds reasonable.

Tom Freund: Sounds reasonable.

Andy: As Mark's co-editor, I agree.

ACTION: Editors to produce updated Working Drafts by 7th

 

ACTION: Eric and Ian: create agenda for F2F by 7th.

 

8. Final logistics for F2F in RTP on 14/15 Mar

Eric: Do we need the conference call for the 9th?  I suggest not.

 

Colleen: We should keep it on the schedule for now.

 

Eric: Okay.

 

Ian:  Unless we make an announcement, we will have the meeting, unless we explicitly cancel it, by March 6th.

ACTION: Eric and Ian to send a message confirming or canceling the March 9 teleconference meeting by the 6th.

 

Alastair:  What criteria will be used to make the decision?

 

Eric: Whether or not there is something to be discussed.

 

Paul C: The chairs should table (present) a draft agenda for the F2F at the March 9th meeting.

 

Alastair: I think there will be something to discuss, so we should plan to have the meeting.

 

Ian: Everyone in the TC should vote to indicate whether they are coming to the F2F meeting.

(http://www.oasis-open.org/apps/org/workgroup/ws-tx/ballot.php?id=915)

 

Ian: I would like to hear if people think they are probably coming, even if they don't finally come.

Ian: Everybody who has not voted on the ballot, please vote.

Ian: 20 are on the ballot as attending now.

 

9. Issue resolution

i018-21 Clarify subordinated CreateCoordinationContext behaviour

  - Consider action #14

Max: The proposal is posted at: http://lists.oasis-open.org/archives/ws-tx/200602/msg00015.html

It has a word document attached: http://lists.oasis-open.org/archives/ws-tx/200602/doc00003.doc

The document has change markings, beginning in section 3, line 182.

There are five changes. [describes them]

 

Eric: Can we adopt these changes?

 

Alastair: Move we adopt the changes.

Peter Furniss: Second

No objections.

The proposal is adopted.

Issues

 

Alastair: Max and I have considered some other changes, and may raise another issue to clarify CreateCurrentCoordinationContext message.

 

Eric: That will be a separate issue?

 

Alastair: Yes

 

      i015  Replace namespace and action URIs

Ian: it will be useful to resolve issues 15 on this call.  We should have the new namespace URIs in the updated working drafts.  We have a concrete proposal from Colleen regarding the use of the hyphen.  I agree with what Colleen said,  I want to get agreement on the issue of removing the hyphen in WS-TX.  The TC namespace needs to include the TC short name, which is WS-TX.  There is an emerging namespace artifact guideline in ASIS, suggesting the TC short name should not include the hyphen.  I'm proposing that we should go ahead and make a decision on removing the hyphen, so we will not care which way the ASIS group goes.

 

Martin: The ASIS proposal is not likely to be adopted as it is.  That section is somewhat out of date.  It is likely that there will be no change.

 

Ian:  We should go ahead and make a decision.

 

Martin: We should go back and change the TC short name if we remove the hyphen from the namespace.

 

Ian: Colleen's proposal is to keep the hyphen.  It is at http://www.oasis-open.org/archives/ws-tx/200602/msg00007.html

 

Eric: The proposal is to keep the hyphen.

Do we have consensus?  Is there any objection to keeping the hyphen?

 

Ram Jeyaraman:  There are a couple of documents which need reflect the resolution of this issue.

 

Colleen: We can follow what other committees like RX are doing.  There is a link to what they are doing, in my proposal.

 

Ian: We also need to supply values for the yyyy/mm in the namespace.  We should use 2006/03, based on the date we expect the CD or CDs to be approved.  It is not a critical date; we just need to establish it.

 

Ian: Move to adopt the versioning policy including the WS-RX policy.

Ram: Second.

 

Monica: Can someone explain the WS-RX model for versioning?  
 
Ian: The pointer is http://docs.oasis-open.org/ws-rx/issues/ReliableMessagingIssues.xml#i088

 

Ram: The policy states that the namespace will not change arbitrarily.  It will change only when a published CD or specification results in a non-backwardly compatible namespace, and defines the changes which are considered not to be backwardly compatible.

 

Alastair: It should say either CD or specification changes.

Eric: Yes, that is included.

 

Martin: I would like to feed our decision back to the ASIS review.

ACTION: Chairs to inform ASIS that we are not following the guideline with respect to hyphens.

 

Eric: Any objections to passing the proposal, including the WS-RX policy?

No objections noted.

Proposal for resolution of issue 15 has passed.

 

i009  Is request-reply MEP useful?

Last message on issue 009:  http://lists.oasis-open.org/archives/ws-tx/200601/msg00014.html

 

Alastair:   This is why I raised the point in the minutes of 9 February, to highlight that there is no reason to use the request-reply MEP in the specification.  It is not necessary.  I asked why it exists.  I believe it is an unnecessary complication.  The discussion in section 9 of WS-AT on header elements would be moved to WS-Coordination and be used by reference in WS-AT and WS-BA.  I also wanted to be more precise in the terminal or non-terminal message.  I have proposed some modification to the text to make it more useable as a reference from WS-AT and WS-BA.

 

Eric: So there are two parts to your proposal

 

Alastair: The material in AT is relevant to BA.  It is better to move it to WS-Coordination and use it in both specs.

The terminal/non-terminal discussion needs to be changed.

 

Martin:  I don't object to moving the text to WS-Coordination.    I don't see the need to change the MEPs.

 

Max: Agree with Martin.  I also agree that Alastair's stating it as two parts, editorial and substantive, is right.

The issues in coordination should be in WS-C, but it may be better to have specific clarifications also in WS-AT and WS-BA.

In the original specification, the MEP choice was intended to be the most natural way to accomplish each part, not to make it all Request-Response, or all one-way.  It is not broken, no need to change it.  There is no compelling or strong motivation.  We should keep it request-reply.

 

Eric: Summarizing: There is a possibility of separating editorial from technical change.  There appears to be divided opinion on the technical changes.  Everyone who has spoken apparently agrees with the editorial change [moving text to WS-Coordination].

 

Alastair: Yes, separate the editorial and technical aspects.  I don't think it is good practice to have multiple ways to do something.  It is better to use one MEP for this.  It is underspecified.  I posted 3 documents including "gaps in the status quo" which discuss this. 

If you are going to continue with the Request-Response MEP, there should be a statement with respect to how the bounds are handled.  If RR is maintained by the TC, I will be happy to submit an issue to clarify this usage.

 

Eric: should we vote separately on this?  There is support for making the editorial change to consolidate in WS-C.  Can we break these out as separate issues?

 

Alastair: I'll be happy to not force a vote on retaining RR. 

 

Eric: I would like to have issue 9 resubmitted to reflect this.

 

Ian: maybe we can be more efficient.  If there is a concrete proposal maintaining the RR MEP, we could address that separately.  I think the editors of the WS-AT and WS-C specs could get together and produce a concrete proposal.

 

Colleen:  we should have a concrete proposal.

 

Alastair: I will try to address this, as long as we can discuss it on the 9th.

 

ACTION: Alastair, editors of AT and C and BA to develop concrete proposal to move material from section 9 of WS-AT on header elements to WS-Coordination and allow it be used by reference in WS-AT and WS-BA.

 

Resolution: Issue 9 remains open.

 

Tom Rutt: If one needs reliability why not compose ws-rm or ws-reliability with ws-tx?

 

i016  ReplaceParticipant

Not discussed.

i022 WS-C: Clarify normative requirements for MU attribute

Not discussed.

 

10. AOB

None

 

11. Straggler Roll Call

Conducted by Eric.

 

Final attendance, from the WS-TX web page:

 

Meeting Attendees

Name                            Company                      Status

Frank Ryan                   Active Endpoints, Inc.*   Group Member

Charlton Barreto             Adobe Systems*            Group Member

Peter Furniss                 Choreology Ltd*             Group Member

Alastair Green               Choreology Ltd*             Group Member

Tom Rutt                       Fujitsu Limited*             Group Member

Robert Freund                Hitachi, Ltd.*                 Group Member

Doug Davis                    IBM*                             Group Member

Thomas Freund             IBM*                             Group Member

Daniel House                 IBM*                             Group Member

Ian Robinson                 IBM*                             Group Member

Andrew Wilkinson          IBM*                             Group Member

John Harby                    Individual                       Group Member

Eric Newcomer             IONA Technologies*       Group Member

Paul Cotton                   Microsoft Corporation*    Group Member

Colleen Evans                Microsoft Corporation*    Group Member

Max Feingold                 Microsoft Corporation*    Group Member

Marc Goodner                Microsoft Corporation*    Group Member

Ram Jeyaraman             Microsoft Corporation*    Group Member

Jonathan Marsh             Microsoft Corporation*    Group Member

Asir Vedamuthu             Microsoft Corporation*    Group Member

Paul Knight                   Nortel Networks Limited* Group Member

Martin Chapman            Oracle Corporation*        Group Member

Anish Karmarkar            Oracle Corporation*        Group Member

jeff mischkinsky             Oracle Corporation*        Group Member

Alex Yiu                        Oracle Corporation*        Group Member

Alain Regnier                 Ricoh Company, Ltd.*    Group Member

Joseph Fialli                  Sun Microsystems*        Group Member

Monica Martin                Sun Microsystems*        Group Member

Mohamed Aariff             Tibco Software Inc.*       Group Member

Shivajee Samdarshi        Tibco Software Inc.*       Group Member