W3C

Web Services Addressing Working Group Teleconference

19 Dec 2005

Agenda

See also: IRC log

Attendees

Present
Francisco Curbera (IBM Corporation)
Vikas Deolaliker (Sonoa Systems, Inc.)
Paul Downey (BT)
Robert Freund (Hitachi, Ltd.)
Marc Hadley (Sun Microsystems, Inc.)
David Hull (TIBCO Software, Inc.)
Yin-Leng Husband (HP)
David Illsley (IBM Corporation)
Anish Karmarkar (Oracle Corporation)
Paul Knight (Nortel Networks)
Philippe Le Hégaret (W3C)
Mark Little (JBoss Inc.)
Jonathan Marsh (Microsoft Corporation)
Jeff Mischkinsky (Oracle Corporation)
Nilo Mitra (ERICSSON)
David Orchard (BEA Systems, Inc.)
Gilbert Pilz (BEA Systems, Inc.)
Tony Rogers (Computer Associates)
Tom Rutt (Fujitsu Limited)
Katy Warr (IBM Corporation)
Pete Wenzel (Sun Microsystems, Inc.)
Ümit Yalçınalp (SAP AG)
Prasad Yendluri (webMethods, Inc.)
Absent
Abbie Barbir (Nortel Networks)
Andreas Bjärlestam (ERICSSON)
Dave Chappell (Sonic Software)
Eran Chinthaka (WSO2)
Glen Daniels (Sonic Software)
Jacques Durand (Fujitsu Limited)
Michael Eder (Nokia)
Marc Goodner (Microsoft Corporation)
Arun Gupta (Sun Microsystems, Inc.)
Hugo Haas (W3C)
Amelia Lewis (TIBCO Software, Inc.)
Bozhong Lin (IONA Technologies, Inc.)
Eisaku Nishiyama (Hitachi, Ltd.)
Ales Novy (Systinet Inc.)
Rich Salz (DataPower Technology, Inc.)
Davanum Srinivas (WSO2)
Jiri Tejkl (Systinet Inc.)
Steve Vinoski (IONA Technologies, Inc.)
Regrets
Mike Vernal (Microsoft Corporation)
Steve Winkler (SAP AG)
Chair
Mark Nottingham
Scribe
Prasad Yendluri

Contents


<scribe> scribe: prasad

2. Agenda review, AOB

Mark review agenda

3. Call for corrections to the minutes

No corrections

resolution: approved minutes of dec 12 05

4. Review action items

all pending

XMLP Request for requirements feedback

Anish: Issue is still open
... issue to resolve in soap 1.2 erratum or in the new XMLP work

<Katy> np

Anish: We initially did not want to fix in errata as it impacts too much in the spec. But Noah may have changed mind

DHull: I was on the XMLP call. Noah thinks can be done in erratum. Does not see a need for one-way MEP

<RalphS_> sorry for the inconvenience, folks

Umit: When is the decision on errata vs new work going to be made?

Anish: I don't think it is either or

prank call ?

<dorchard> please paste # into irc

<marc> did someone get that number ?

866-214-3176

<Marsh> 1-866-214-3176:

<Marsh> #2979273

<mnot_2> +1 866 214 3176 (US)

<mnot_2> +1 404 827 9098 (non-US)

<mnot_2> Access Code: 2979273

<TRutt> sounds like downs syndrome to me

swithing over to other line

<TRutt> Host not yet arrived

Restarting

Trutt: Distingushing reliable vs non-reliable transports is important

DHull: Do we really need one-way MEP from XMLP?

Anish: One-Way MEP in Errata or new effort?

DHull: New

DaveO: Real question is req optional response good enough?

<dorchard> Seems to me that wsdl one-way could map to soap request-optional-response mep.

<dorchard> And I'd rather not "hijack" this discussion to meet some other requirement.

<dhull> cheese sandwich with no bread

Umit: Is whats on the table maping WSDL MEP to SOAP optional resp

dhull: believe so, not sure

mnot: No feed back at this pint to XMLP. Moving on

DHull: We can set the expectation as our feedback

6. Test Suite and Testing Update

Mnot: Where are we on that

Paul: Probably on track
... MS put out two end points. IBM put out a web page
... some test broke; trying to fix

Mnot: Plan to do some tests end-to-end for the vancouver f2f

7. Working Draft Issues

i066 - wsaw:UsingAddressing as a policy assertion

i059 - Support for asynchronous / multi-MEP usage of web services

Mnot: We split i59 into three parts last week

<dorchard> For the record, I haven't had a chance to review the umit's proposal because it arrived on Sunday night.

Mnot: i67 and i68 in addtion to i59

<mnot_2> http://www.w3.org/mid/[email protected]

<mnot_2> 

Mnot: Umit has a new proposal
... Too many Proposals and status is confusing

Marsh: Proposal does not seem specific enough
... If I have a method tht has a reply and fault and one of them is anonymous am i required to generate a fault?

<dorchard> Seems to me that Jonathan's concerns about fault handling might need to be handled by a new issue.

<dhull> fault goes back as transport-level response (if available). After we understand WSA headers and WSA is engaged, anonymous means "response channel"

<dhull> +1 to DaveO

<dhull> prohibited is currently defined in terms of what may appear (not what is used)

<anish> +1 to daveo

<dhull> +1 to anish

<anish> difference between generating a fault and sending a fault

discussion on malformed fault handling

scribe: between umit, Marsh and others

<dhull> Let's not say "anonymous" so much here. Back-channel, response or whatever makes sense outside of WSA. "Anonymous" only means anything in WSA.

<pauld> Agenda: http://lists.w3.org/Archives/Public/public-ws-addressing/2005Dec/0079.html

Marc: We provide two ways to do using addressing and in SOAP module we don not proide a way to do anonymous equivalent

<anish> you could use a property

Mnot: Can do discuss this in i67?

DaveO: U can set properties but, properties are global and not scoped to an operation

<Zakim> dorchard, you wanted to mention that SOAP already has this problem, and it says that faults may or may not be delivered... c'est la vie.

<uyalcina> it is not as is. You can always raise issues about the wording.

<mnot> ACTION: Umit to incorporate anon element into example 3-3 in conjunction with wsoap:module [recorded in http://www.w3.org/2005/12/19-ws-Addr-minutes.html#action01]

Anish: asks for examples in section 3-3 use of anonymous element in conjunction with SOAP module

Marc: Concerned with including text in section 3.4

MNot: None questioned 3.4 so far

<anish> http://lists.w3.org/Archives/Public/public-ws-addressing/2005Dec/att-0080/ws-addr-wsdlProposedRevision1.62.html

Mnot: Like to close i59 with Umit's proposal with the "anonymous text" in 3.3 deleted and examples added, text in section 3.4 is subject to editorial update.

Marc: 3.4 is not going in the doc?

MNot: It is, but can raise issues againt it.

Umit: Why don't we add an editoril note in section 3.4 that the folllowing text is subject to change?

<dorchard> Why not just accept up to 3.4? 3.4 is easily talked about in email/proposals.

<dhull> 3.4.1 captures one useful scenario. OK with noting that, but there are useful scenarios that conflict with that text. OK with 3.4.1 as long as it's clearly scoped

<bob> +1 Katy

Katy: Likes Umit's proposal

MArc: uncomfortable with putting 3.4 in spec

Mnot: That is a separate issue

<uyalcina> Not accepting 3.4 is mixed with the concept of putting it to a different document. This is not a content issue.

TRutt: I would go with putting in 3.4 with a disclaimer

Anish: Adding a note seems a reasonable way fwd

Marsh: When there is no consensus what is the status quo? What is in the doc ?
... What we have not agreed to should not be in the doc

<pauld> +1 to Jonathan - the document should reflect the Status Quo, not least for generating test cases

<Zakim> dhull, you wanted to clarify discomfort with 3.4.1

<dorchard> +1 to Jonathan on declaring victory up to 3.4

<Zakim> anish, you wanted to ask about "empty SOAP envelope in section 3.4

Anish: In 3.4.1 2nd sub-bullet 1. What is "empty SOAP Env"? Empty HTTP entity Body, with SOAP hdrs ?

Marc / Umit: we need another issue on this

Umit: Options: Don't say anything about SOAP 1.1 HTTP binding; put text with disclaimers..
... I am infavor of putting in w/ disclaimer

Marc: Can we find in the minutes what we agreed to?

Paco: What we agreed to at the f2f is status quo

Trutt: Text in 3.4 reflects what we agreed in the meeting

<dhull> In other words, the rule is, if you get a message over HTTP, you *must* send something back (which is what HTTP demands anyway). If you don't know what to say, send an empty 202.

MNot: proposes we accept as resolution for i59 all changes in Umit's proposal upto section 3.2 and editor's recollection of what we agreed to at f2f ...

Anish: what about examples?

Mnot: Those included

<dorchard> Marc, Mark said "upto section 3.4 "

Mnot: Any objections to proposal?

Marsh: I object

Taking a formal vote

<yinleng> abstain

<dorchard> why are people abstaining?

<dorchard> Is it because not enough text going in? Or too much?

Yes - 10; No -2; Abs - 3;

Mnot: Motion Carries; i59 is closed

No meeting next 2 weeks. Reconvene Jan 9th

Adjurn

Summary of Action Items

[NEW] ACTION: Umit to incorporate anon element into example 3-3 in conjunction with wsoap:module [recorded in http://www.w3.org/2005/12/19-ws-Addr-minutes.html#action01]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.127 (CVS log)
$Date: 2005/12/20 02:05:50 $