Minutes of XMLP WG telcon, 27 November 2002
Present
BEA Systems David Orchard
Canon Herve Ruellan
DaimlerChrysler R. & Tech Mario Jeckle
IBM David Fallside
Macromedia Glen Daniels
Microsoft Corporation Henrik Nielsen
Oracle Anish Karmarkar
SAP AG Gerd Hoelzing
SeeBeyond Pete Wenzel
Software AG Michael Champion (scribe)
Sun Microsystems Marc Hadley
Systinet (IDOOX) Jacek Kopecky
W3C Yves Lafon
W3C Carine Bournez
Excused
Canon Jean-Jacques Moreau
DaimlerChrysler R. & Tech Andreas Riegg
Microsoft Corporation Martin Gudgin
Oracle Jeff Mischkinsky
SAP AG Volker Wiechers
Software AG Dietmar Gaertner
Systinet (IDOOX) Miroslav Simek
Regrets
AT&T Mark Jones
Ericsson Nilo Mitra
IBM John Ibbotson
IBM Noah Mendelsohn
Matsushita Electric Ryuji Inoue
Progress Software Colleen Evans
Tibco Don Mullen
Absent
AT&T Michah Lerner
eXcelon Jim d'Augustine
Fujitsu Limited Kazunori Iwasa
Fujitsu Limited Masahiko Narita
IONA Technologies Oisin Hurley
IONA Technologies Eric Newcomer
Martsoft Jin Yu
Mitre Marwan Sabbouh
Mitre Paul Denning
Netscape Vidur Apparao
Netscape Ray Whitmer
Tibco Amy Lewis
Unisys Lynne Thompson
Unisys Nick Smilonich
[davidF] zakim, who is here?
[Zakim] On the phone I see Marc, +49.622.774.aabb, GlenD, ??P3, Henrik?, ??P5, Fallside, Carine, ??P7, Yves, JacekK (muted), Mike_Champion
[Zakim] On IRC I see mchampion, JacekK, Zakim, RRSAgent, GlenD, marc, davidF, jjm-away, Henrik, caribou, Loggy, Yves
[Zakim] +DOrchard
[anish] anish has joined #xmlprotocol
[Yves] /nick scribe
[DaveO] DaveO has joined #xmlprotocol
[scribe] Minor revisions to published agenda were noted
[scribe] See latest message from chair
[scribe] Agenda item 3: Approval of mintutes of last telcon.
[scribe] Request to postpone accepted.
[scribe] F2F day 2 minutes still pending
====================
[scribe] Action items
[scribe] Chair: First still pending,
[scribe] PaulD: matchup of req doc to spec, no sign of it.
[scribe] Volunteer - Mark will do it.
[Henrik] ACTION: MarcH: Match latest spec against requirement sec #4.4, 4.5 ... by 8/11 (taken over from PaulD)
[scribe] Oops, "Marc will do it"
[scribe] Editors: cleanup some language. Thinks they have made necessary changes. Action marked as done
[scribe] Jean-Jacques: Media type registration done
[scribe] Editors: done
[scribe] Yves: DIME authors' permission ... still pending. Working on it!
[scribe] Yves: IPR page. pending. All replies have been received.
[scribe] DavidF: Resolution of abstract model doc. Pending
[scribe] Colleeen: AM issues Done.
[scribe] DavidF: Issues ... done
[scribe] DavidF: Attachment feature doc editors - Pending.
=====================
[scribe] STATUS REPORTS
[scribe] Primer - Milo sent report. no questions.
[caribou] s/Milo/Nilo
[scribe] Spec: 3 remaining issues, all on agenda. 395, MEP assumptions about intermediaries, Soap features/modules relationship. No other editorial work this week
[scribe] Last call issues list: 3 issues left
[scribe] TAG IPR situation: Yves got final replies from MS. Needs to verify that it is complete. Will update IPR page and make it public.
[scribe] We will not have a PAG.
[scribe] The new list will note all members in good standing, time that former members left the WG.
[scribe] Questions about IPR?
[scribe] none.
[scribe] Implementation Tracking: Implementers met twice to get things wrapped up. Now have complete set of features and tests.
[scribe] Table 2 has content for everything. DavidF has action to update table with results from last implementers telcon.
[scribe] All implementers contacted have said that my next week's telcon they can update their bits in the feature list and Table 2
[scribe] Posed question of when implementers would have done live tests for all features.
[scribe] Estimates ranged from 3 to 12 weeks
[scribe] Don't need all tests for all implementations done, just 1 pair of implementations that implement all features.
[JacekK] rather for each feature one pair of implementations (no implementation need implement everything)
[scribe] Probably 4-6 weeks to be done. Early January to have evidence that all features are testable.
[scribe] Jacek agrees, switch is already turned on and we are accumulating checkboxes.
[scribe] Clarification: For every feature we need a pair of implementations; no single implementation has to implement all features.
[scribe] Do we need a formal CR period? It might be a good idea.
[scribe] This would give us a landmark (milestone?) to shoot for.
[scribe] CR may not produce any changes to CR spec, but it would help get closure, and wouldn't be huge
[scribe] Marc: would be in favor. Many questions about where we are in the process
[scribe] Jacek: agrees
[scribe] How long will it take? Chair can write any number he wants in there!
[scribe] Having "Candidate Recommendation" out is a Good Thing. Shoot for having it out in a couple of weeks, month length.
[scribe] Yves: mid-January OK. No mention of minimum CR period in process doc. 3 weeks is a practical minimum.
[scribe] What will rest of the world think of this? We don't want people thinking they can tweak on it!
[scribe] CR is for IMPLEMENTATION EXPERIENCE, so we have a good reason to push back on other types of feedback.
[scribe] CR is NOT a general review period.
[scribe] TAG has SOAP on the agenda next week re XML subset. Do we make an exception for them?
[scribe] Informal understanding that TAG will push back on this issue on procedural grounds. Paul Grosso can raise it directly with us.
[scribe] DaveO: Let's worry about this if the TAG raises it with us, don't worry about it until then.
[scribe] Chair: this would be the third time the TAG has raised such issues and delayed us, and this one is very late.
[scribe] Requests for clarification would be received with more friendliness than requests for changes.
[scribe] No objections to having a CR period.
[scribe] ACTION: chair and staff will make plans for CR period
[scribe] Anish: concerns with implementer group and test collection doc.
[scribe] Our doc is for interop, not compliance. Questions whether we have achieved our goal if the implmementers are not using the test collection doc as intended.
[scribe] If not, why update it?
[scribe] Chair: email from Don Mullen about this. Others on call wish to respond? How do we address it in test collection doc?
[scribe] Henrik: Don has good ideas, defers to them
[DavidCF] don's url is http://lists.w3.org/Archives/Member/w3c-xml-protocol-wg/2002Nov/0072.html
[scribe] Don's suggestions: Fix assertion test numbering so that they are stable;
[scribe] Table 2 should reference doc
[scribe] Tests should reference soapbuilders tests
[scribe] Test collection document used to document new tests from implementers.
[scribe] Sense of WG on these suggestions?
[scribe] Anish: likes general direction.
[scribe] Chair: let's not go into detail. Suggests that Anish respond to Don's email.
[scribe] ACTION: Anish to respond to Don, WG should follow this thread closely.
=========================
[scribe] ISSUES LIST:
[scribe] Can header blocks exist outside modules? "Final" version of agreed text got comments back.
[scribe] JJ not on call ... seeks another participant to summarize ...
[scribe] Overview from Henrik - specifies relationship between features and modules, has mundane edits.
[scribe] Discussion of whether module is feature at F2F, concluded that modules are different from features. Many attempts to clarify now in text.
[scribe] Now have different guidelines for describing features and modules.
[scribe] Any objection to us making the changes proposed in JJ and Glenn's email. None.
[scribe] Editors will "do right thing" with grammar.
[Henrik] ACTION: Editors: Incorporate text for SOAP features vs. SOAP modules as described in http://lists.w3.org/Archives/Public/xml-dist-app/2002Nov/0081.html (and thread)
[DavidCF] re. previous action - the editor's "right thing" - s/and message exchange patterns (MEPs) such/or message exchange patterns (MEPs) such/
[scribe] MEP Assumptions about intermediaries issue --
[scribe] Chair: No email discussion, so we must revisit it here.
[scribe] Henrik's summary -- has been discussed 3 times. Does MEP work for intermediaries or not? Proposal is to clarify that 2 MEPs that we have DO NOT apply to intermediaries.
[scribe] There has been discussion with respect to whether intermediaries can be opaque with respect to MEP.
[scribe] MEPs that we define are hub to hub. We are open to extensions in all sorts of ways.
[JacekK] s/hub/hop/g
[scribe] Glenn: Discussion last week worrying about closing off intermediaries that are invisible.
[scribe] They should not be DISallowed. That is too limiting. How do MEPs compose? Are there different end to end and hub to hub MEPs?
[scribe] WSDL is wrestling with this ...
[scribe] Not enough real work has been done on this ... would be good use of CR period.
[scribe] Henrik: it will take 5 years to learn this. We could say that MEP is designed for single hub, MAY apply to other situations but we don't assert that.
[scribe] Glenn: Thinks this will satisfy his concern.
[scribe] Henrik: We will modify "no intermediaries" assertion to say that it has been designed without reference to intermediaries.
[scribe] Glenn: remove sentence, it is implied elsewhere. It is confusing
[scribe] Henrik: can live with this. Problem is that someone raised issue of which it is.
[scribe] Chair: We were trying to clarify for benefit of people who did ask question.
[scribe] Chair: Can we assign Henrik or Glenn to write the necessary sentence, and we'll decide. They agree.
[Henrik] ACTION: GlenD: Provide sentence as a note that we have nothing to say about the presense of intermediaries in the context of the two MEPs provided by the SOAP 1.2 spec
[scribe] Henrik: can we say that we accept rest of clarifications and wait for note?
[scribe] Glen: Asks about "is not a goal" wording ...
[scribe] Henrik: that is another issue, out of scope for this discussion.
[DavidCF] original proposal is at http://lists.w3.org/Archives/Public/xml-dist-app/2002Nov/0005.html
[scribe] Glen: agrees with Henrik's proposal.
[Henrik] with the exception of the sentence: "The MEP does not support the presence of SOAP intermediaries
[Henrik] in the message path. "
[DavidCF] proposal is to take (A) and (B) without the sentences "The MEP does not support the presence of SOAP intermediaries
24 [DavidCF] in the message path"
[Henrik] ACTION: Editors Incorporate resolution stated in http://lists.w3.org/Archives/Public/xml-dist-app/2002Nov/0005.html as stated in A) and B) without the sentences "The MEP does not support the presence of SOAP intermediaries in the message path"
[scribe] Marc: Encoding style ... this is purely editorial, text says it can go on SOAP header block ....
[scribe] Intended to say that it couldnt go on elements we define, but hole in the way we formulate rules.
[scribe] Current text is misleading, suggests that a SOAP fault element could have an encoding style.
[scribe] Henrik: schema says this already.
[scribe] Jacek: what if we stick an update header into body? Does encoding style fit there?
[scribe] Marc: can't put encoding style on those things. Text is not clear about this.
[scribe] Proposal to add to bullet item 2 adn to section 5 to clarify.
[scribe] Jacek has not proposed text yet. Agrees with Marc's proposal.
[scribe] Chair: do we accept change? No objection.
[DavidCF] http://lists.w3.org/Archives/Public/xml-dist-app/2002Nov/0045.html
[DavidCF] ... is url for marc's prposal
[Henrik] ACTION: Editors: Incorporate suggestion from MarcH as stated in http://lists.w3.org/Archives/Public/xml-dist-app/2002Nov/0045.html
[scribe] Marc will put in sentence in text that defines Body / Child elements that we will explicitly disallow soap-encoding attributes.
[scribe] This is in addition to Jacek's comments in message 47
[DavidCF] whose url is http://lists.w3.org/Archives/Public/xml-dist-app/2002Nov/0047.html
[Henrik] ACTION: Editors have the permission to clarify editorially the question raised in http://lists.w3.org/Archives/Public/xml-dist-app/2002Nov/0047.html
[scribe] ISSUE - raised by Glen about QNames and URIs, never posted comment.
[scribe] Told chair that he is thinking about removing it. Chair will contact him and clarify. Topic postponed.
[DavidCF] ACTION: chair contact glen re status of QNAME/URI topic listed in agenda (with a view to removing it)
[scribe] ISSUE: Jacek offered clarification to Noah's request for clarification ...
[Iwasa] Iwasa has joined #xmlprotocol
[scribe] Jacek: Noah concerned with encoding with respect to simple and compound values, missing values.
[scribe] Problem is that "value" is used in different ways in different places in spec. Confusing, but not serious issue. Editors may want to clean this up if they have time.
[scribe] Henrik: issue was to remove word "optional"
[scribe] Jacek agrees. That is the proposal. Just remove word "optional".
[scribe] No objections heard.
[Henrik] ACTION: Editors: Incorporate clarification to resolving issue 389 as stated in http://lists.w3.org/Archives/Public/xml-dist-app/2002Nov/0042.html (i.e. just remove the word 'optional')
[scribe] ISSUE Marc raised question about xml:lang and fact that we thought there would be unique values of this attribute.
[scribe] Text does not reflect fact that these xml:lang values should be different ... or MUST be different.
[scribe] Consensus seems to be that SHOULD is the appropriate wording. It's a nice to have, not an interoperability issue.
[scribe] Relates to human readers, not machines.
[scribe] Does anyone disagree. None heard.
[Henrik] ACTION: Editors: Clarify that the values of the xml:lang attribute in the Reason EII SHOULD NOT be duplicated as per issue raised by MarcH http://lists.w3.org/Archives/Public/xml-dist-app/2002Nov/0055.html
[scribe] ISSUE 395 - Came up at F2F to see that HTTP checks that there is no DTD.
[scribe] Motivation is that we removed text in Part 1 and we wanted to make sure that this was prohibited.
[scribe] Henrik: We don't know what other media types specify. In soap+xml we can make it very clear what we accept.
[scribe] Proposal is to do nothing. No dissent.
[Henrik] ACTION: Henrik: Send closing text (including rationale) for issue 395 as stated in http://lists.w3.org/Archives/Public/xml-dist-app/2002Nov/0078.html
[scribe] ACTION: Henrik will send appropriate message to xmlp-comments with his rationale on issue 395
[DavidCF] url for colleens AM comments http://lists.w3.org/Archives/Member/w3c-xml-protocol-wg/2002Nov/0061.html
[scribe] Colleen's proposal for abstract model issues.
[scribe] The Abstract Model WD will be parked in "working drafts no longer under development"
[scribe] Colleen proposes resolution text. SOAP meets R300 requirement, doesn't mention applicaiton semantics, and we won't further refine it.
[scribe] Does this resolve issue 34? No objections.
[scribe] Issue 86: This issue is moot because it pertains to AM doc. No objections to closing this issue with proposed text.
[scribe] Issue 87: Asked to clarify a parameter; same proposed resolution as 86.
[scribe] No objection to closing 87
[scribe] Issue 89. Similar. No objection to closing 89.
[scribe] Issue 90: Unit data. Construct that pre-dated MEPs. Proposed resolution is to note that this is no longer in the main spec, AM document is moot, so issue closed. No objection.
[scribe] Issue 96: Asked to reconcile AM and main spec requirements. Reslution same as others here. No objection to closing 96.
[JacekK] ACTION: DaveO to send resolution text on issues 86, 87, 89, 90, 96 to xmlp-comments
[scribe] Issue 385 postponed.
[scribe] Compound message raised. Is a compound message with zero attachments "compound". Noah provided response. Do we do anything about it, since it came in after last call?
[Henrik] Noah's informal response: http://lists.w3.org/Archives/Public/xml-dist-app/2002Nov/0039.html
[caribou] does action 14 include issue 34 as well?
[scribe] No further commentary. No need for further action by us.
[scribe] DavidF: Are there any other issues that we need to think about here? None.
[scribe] Agenda item 7 - IANA application for our media type pieces in place.
[scribe] We will submit application when we go to PR. No action needed at moment. Is this OK with everyone? Other issues to consider? None raised.
[scribe] Agenda Item 8 - Marc, C14N discussed at F2F. Intermediaries can cause
[scribe] problems for digital signatures. Marc and Martin G put together extension of a canonicalizaiton alorithm can considers transforms that a SOAP intermediary could do.
[scribe] We could submit that as a Note to W3C, or WG could put it into spec.
[scribe] Proposal to WG to take on the ownership of this rather than individual companies submit it.
[scribe] Chair: is this pre-SOAP 1.2 spec? Marc - assume that is would be done in parallel.
[scribe] Treat it same as attachment spec. On its own time frame.
[scribe] What is priority? Marc- Lower than attachments spec, probably.
[scribe] Henrik: maybe it should be a Note soon, then we figure out whether to pick it up.
[scribe] Chair: may be premature for WG since we haven't looked at it closely. Suggests that Marc ping WG, we'll discuss next weeh
[Henrik] ACTION: MarcH: Send a mail to WG-only list pointing to the c14n document that he and Gudge have been working on
[scribe] WSD WG asked us to provide coments on last call requirements doc by end of December. DavidF seeks volunteer to draft comments. No one on telcon volunteers.
[Henrik] ACTION: DF: Ask for volunteers to review WSD requirements document and generate feedback. Due date for comments is Dec 31)
[scribe] XMLP WG also asked by WS CG to consider what we will do after SOAP 1.2 besides attachment document.
[scribe] DavidF: there have been few recent discussions on this. Perhaps because WS-Arch now has mandate to recommend new WGs.
[scribe] Henrik: can we just go home?
[scribe] Chair: motivation is to preserve the WG that has learned to work together for new work.
[scribe] Chair: if people have ideas about this, send to member list.
[Henrik] ACTION: DF: Send mail to WG-only list to kick off discussion about the question "what does XMLP
[scribe] "We'll let WS Arch worry about this" is a satisfactory answer to the CG.
[Yves] rrsagent, excuse us
[RRSAgent] I see 17 open action items:
[RRSAgent] ACTION: MarcH: Match latest spec against requirement sec #4.4, 4.5 ... by 8/11 (taken over from PaulD) [1]
[RRSAgent] recorded in http://www.w3.org/2002/11/27-xmlprotocol-irc#T19-14-50
[RRSAgent] ACTION: chair and staff will make plans for CR period [2]
[RRSAgent] recorded in http://www.w3.org/2002/11/27-xmlprotocol-irc#T19-42-32
[RRSAgent] ACTION: Anish to respond to Don, WG should follow this thread closely. [3]
[RRSAgent] recorded in http://www.w3.org/2002/11/27-xmlprotocol-irc#T19-48-38
[RRSAgent] ACTION: Editors: Incorporate text for SOAP features vs. SOAP modules as described in http://lists.w3.org/Archives/Public/xml-dist-app/2002Nov/0081.html (and thread) [4]
[RRSAgent] recorded in http://www.w3.org/2002/11/27-xmlprotocol-irc#T19-54-34
[RRSAgent] ACTION: GlenD: Provide sentence as a note that we have nothing to say about the presense of intermediaries in the context of the two MEPs provided by the SOAP 1.2 spec [5]
[RRSAgent] recorded in http://www.w3.org/2002/11/27-xmlprotocol-irc#T20-04-16
[RRSAgent] ACTION: Editors Incorporate resolution stated in http://lists.w3.org/Archives/Public/xml-dist-app/2002Nov/0005.html as stated in A) and B) without the sentences "The MEP does not support the presence of SOAP intermediaries in the message path" [6]
[RRSAgent] recorded in http://www.w3.org/2002/11/27-xmlprotocol-irc#T20-08-32
[RRSAgent] ACTION: Editors: Incorporate suggestion from MarcH as stated in http://lists.w3.org/Archives/Public/xml-dist-app/2002Nov/0045.html [7]
[RRSAgent] recorded in http://www.w3.org/2002/11/27-xmlprotocol-irc#T20-13-16
[RRSAgent] ACTION: Editors have the permission to clarify editorially the question raised in http://lists.w3.org/Archives/Public/xml-dist-app/2002Nov/0047.html [8]
[RRSAgent] recorded in http://www.w3.org/2002/11/27-xmlprotocol-irc#T20-16-26-1
[RRSAgent] ACTION: chair contact glen re status of QNAME/URI topic listed in agenda (with a view to removing it) [9]
[RRSAgent] recorded in http://www.w3.org/2002/11/27-xmlprotocol-irc#T20-18-51
[RRSAgent] ACTION: Editors: Incorporate clarification to resolving issue 389 as stated in http://lists.w3.org/Archives/Public/xml-dist-app/2002Nov/0042.html (i.e. just remove the word 'optional') [10]
[RRSAgent] recorded in http://www.w3.org/2002/11/27-xmlprotocol-irc#T20-21-23
[RRSAgent] ACTION: Editors: Clarify that the values of the xml:lang attribute in the Reason EII SHOULD NOT be duplicated as per issue raised by MarcH http://lists.w3.org/Archives/Public/xml-dist-app/2002Nov/0055.html [11]
[RRSAgent] recorded in http://www.w3.org/2002/11/27-xmlprotocol-irc#T20-24-51
[RRSAgent] ACTION: Henrik: Send closing text (including rationale) for issue 395 as stated in http://lists.w3.org/Archives/Public/xml-dist-app/2002Nov/0078.html [12]
[RRSAgent] recorded in http://www.w3.org/2002/11/27-xmlprotocol-irc#T20-27-53
[RRSAgent] ACTION: Henrik will send appropriate message to xmlp-comments with his rationale on issue 395 [13]
[RRSAgent] recorded in http://www.w3.org/2002/11/27-xmlprotocol-irc#T20-28-18
[RRSAgent] ACTION: DaveO to send resolution text on issues 86, 87, 89, 90, 96 to xmlp-comments [14]
[RRSAgent] recorded in http://www.w3.org/2002/11/27-xmlprotocol-irc#T20-38-09
[RRSAgent] ACTION: MarcH: Send a mail to WG-only list pointing to the c14n document that he and Gudge have been working on [15]
[RRSAgent] recorded in http://www.w3.org/2002/11/27-xmlprotocol-irc#T20-53-00
[RRSAgent] ACTION: DF: Ask for volunteers to review WSD requirements document and generate feedback. Due date for comments is Dec 31) [16]
[RRSAgent] recorded in http://www.w3.org/2002/11/27-xmlprotocol-irc#T20-55-35
[RRSAgent] ACTION: DF: Send mail to WG-only list to kick off discussion about the question "what does XMLP WG want/plan to do after SOAP 1.2? [17]
[RRSAgent] recorded in http://www.w3.org/2002/11/27-xmlprotocol-irc#T21-02-37
[Yves] zakim, excuse us