W3C XML Protocol Working Group teleconference, 10 October 2001

Minutes of XML Protocol WG telcon on October 10, 2001.

1. Roll call, scribes for minutes/action items

Present 32/28 Excused Regrets Absent

2. Agenda review, and AOB

   Agenda item #10 has wrong URL it should be:

   - Nov F2F
       - Poll taken to see who can make it (crisis and budget constraints
         might mean people can't/don't want to travel)
       - 23 responded saying they plan to attend
       - Maybe do a video conference at the F2F for the people who can't
         travel.  However, seating in the video facility at the F2F location
         is for only 25 people (hence the poll).
       - Chris Ferris:
           - No network connectivity at the F2F
           - Maybe an analog line will be there
           - Need to start the registration process
       - Paul Cotton:
           - Just did a F2F - Schema and Query
           - Ran a telecon for the entire time
           - One presentation on video w/person a telecon
           - Did have some analog lines
         David: Let's find out what are all the possible connectivity options.
         Noah: What are the quorum rules?
         David: No rules
         Noah: Just want to make sure we have enough people to make it worthwhile.
         David: As in last F2F, did business and reported back to WG leaving
                open the option, at the Chair's discretion, of reopening an issue.

   - Paul Cotton - agent item #7
       Paul Cotton: Doesn't believe David Orchard's proposal accurately 
                    describes the XML Core WG's position, and has not had time to study it closely.
       Gudge:Clarification: DavidO's proposal is "yes xml:base" which is a change from previous (Noah's) proposal which is "no xml:base"
       Noah: Actually not my proposal which was a write-up of PaulC's. The original proposal concerned the requirement that SOAP processors be forced to understand XMLBase.
       PaulC: Prefers original proposal over DavidO's. Can't see any reason for changing actor URIs to relative URIs. The solution isn't modular enough. Wants to defer agenda item until next week.
       David: OK. Item postponed w/approval of WG.


3. Minutes from 3 Oct telcon approved without dissent.


4. Review action items.

     - Paul Cotton: not yet
     - Paul Cotton: void null/void from query - not yet
         - Action taken to write it up in time for the next telecon
     - TBTF: ongoing
     - John Ibbotson: done
     - Chris Ferris:  
         - Sent issue 12 resolution text
         - Larry unhappy with this resolution
         - Resolution is pending
         - David/Chris will discuss off-line how to proceed
     - Marc Hadley:
         - Write up resolution of issue #4
         - More talks with Larry - agreed to open a new issue.
         - Hugo issue 145 opened - XML decls
     - David Orchard:
         - Wrote proposal
         - Done


5. Status reports

   -- Evaluation of Primer draft
        - Nilo:
            - Status: put out draft 3-4 weeks ago
            - TOC is main focus of draft
            - Examples were taken from parts 1 & 2
            - Paul Denning/Highland provided new examples - had editorials
            - Should we make up new examples?
            - Comments received:
                - John Ibbotson: include usage scenarios?
                - Jean-Jacques: would prefer a more gradual approach to SOAP concepts
                - Include advanced uses of SOAP
            - Suggests we should go thru primer in detail
            - What should we do with the samples? Create new ones or use the ones from
              parts 1 and 2?
            - Noah:
                - +1 to notion that primer should have a gradual approach.
                - It is almost too close to the spec.
                - Primer should include a simple usage scenarios.
                - Would prefer if it showed simple features first.
            - Marwan:
                - Overall likes it
                - Would like primer to show what SOAP's value is.
            - David: Perhaps an intro paragraph would help show such value?
            - Marwan: yes
            - Chris: Notational convention section - does it apply?
            - David: Summary of comments
                        - needs a more gradual introduction to concepts
                        - needs more of a justification
            - Marwan: these are different thoughts, and can probably be attained independently
            - Nilo: Perhaps I started with the more complex issues too soon.
                    Maybe a more gradual approach would help people's concerns.
            - David: Proposal: look into ways of providing a ramp up in complexity and 
                     communicate them back to the group.  Talk with Marwan 
                     about some "motivation" text.
            - Marwan: +1 and will help
            - Nilo: should this discussion take place dist-app?
            - David: Yes. To W3C staff: Include pointer to the Primer on the public page + a mail
                     telling dist-app about it.
   -- Usage Scenarios
        - John Ibbotson:
            - Draft scenario was sent out to WG mailing list
            - Sent a note to Nilo regarding usage scenarios/primer examples positioning
              - Nilo: Sees a problem with combining Primer + usage scenarios.
                      Usage scenario formats (AM model diagrams) are too obscure for target
            - Proposal:
                - We should continue scenario work for internal work using AM
                - Work with Nilo to take the scenarios to rework them so they
                  fit in better with the primer's audience.
            - Dug: Good idea
            - Nilo: Good idea
            - David: Do it.  Need to solicit feedback from WG on usage 
                     scenarios.  What is their status?
            - John: More needs to be documented - new draft by Monday.
            - David: WG should review what has been provided so far and provide
                     John feedback.
            - John: There are many ways to implement them, so if there's a 
                    better idea, let me know.

   -- ETF
        - David: Still attempting to schedule first telcon (will send a schedule email later today), ETF has started compiling list of ETF-related issues

   -- Conformance work report
        - Hugo: Waiting for Oisin for a list of testable assertions.  Will ask for input/contributions from public.
        - Paul Cotton: Have you closed the loop with MS yet?
        - Hugo: Not yet - waiting for a mailing list to be set-up first.

   -- TBTF Task Force
        - Stuart: Status:
                  - Met on Friday
                  - Discussed Framework doc put out on Thursday
                  - TBTF mainly supportive
                           - pro's: framework, rigorous, error codes
                           - con's: too complex, unclear how to test for comformance
                  - TBTF will continue on this path and put it out on dist-app
                  - working with Glen on new drafts
                  - Stuart+Chris will work on updating framework text
                  - MEP needs more work
                  - Other bindings might be coming soon
                  - David asked for discussion from WG
                  - Encouraged WG at large to provide feedback
        - Henrik: Expressed concern about the feasability/robustness of the HTTP binding
        - Chris: Is it too robust or not robust enough?
        - Henrik: Not enough
        - Chris: Talk about other bindings - but who's gonna do it?
                 We need stuff we can actually work with?
        - Paul Cotton: What is the TF asking?  They don't have consensus but
                       they're asking for the WG to do work?  When are we going
                       to decide that the TF has done their job?
        - Noah: TF has tried to do the right thing - balancing all input.
                Presentation could be improved but overall TF likes the 
                Being presented to WG to know if they're WAY off base.
        - David: TF put this forward as a "here's where we're at" some feedback
                 is invited.
                 TF knows it has more work to do.
                 Based on feedback TF knows it needs another instance of a binding.
                 This is not a "formal" request for evaluation, just an 
                 informal checkpoint.
                 Discussion still wasn't made as to whether to distribute it 
                 more widely.
        - Henrik: Shouldn't this discussion happen on dist-app?
        - David: Yes - ultimately but decision has not been made yet.
                 Distinction between WG and dist-app mailing list is still being made.
        - Glen: Move conversation to dist-app ASAP as soon as it is presentable.
        - Paul: When are we going to adopt this as the solution for doing a
                binding?  Can we really leave it open - waiting for feedback.
        - David: yes.
        - Paul: This is not wise it might lead to an irreversible decision.
                I don't want to set the precedence set to have the TF make
                decisions for the WG.  WG should vote if this is "the" draft
                before we go public.
        - Noah: We agreed to do our work in public so why is this an issue?
        - Paul: We're agreeing.  Order of decision making is TF then WG.
                WG needs to decide before we go public.
        - David: So, is the WG ok with making it public and continuing discussion
                 on dist-app?
        - Stuart: It would hurt if people undermined the solution in public.
        - David: I believe we have general consensus on the solution in the WG.
        - Paul Denning: I think it is ready to go if we fill-in placeholders.
        - Paul Cotton: To be clear - if this goes out (on dist-app) MS will argue against it.
        - Henrik: 2 docs - Binding framework - on right track
                         - HTTP binding - on wrong track
        - Hugo: We don't need consensus to publish drafts or to send e-mails
        - Paul Cotton: We need consensus that we need to publish
        - Chris: We're not publishing - we're just putting it on the mailing list
        - Paul Cotton: Understood - I was just quoting Hugo
        - Hugo: It's just an e-mail
        - David: Statement for confirmation: If this is put on dist-app, MS will argue against
its direction.
        - Paul Cotton: Yes, we're going to disagree with the proposal.
                       We need to make sure the decision to put out the doc
                       is a consensus decision.
        - Noah: Consensus is nice but not needed.  TF is self selected and 
                consensus is not needed within TF or even WG.  Public input
                would be good.
        - David: Is WG ok with making it public?
        - Paul Cotton: Iam Ok with making it public now people know we're going to
                       argue strongly against it.
        - David: Action: We'll make it public.


6. XML Document Type

     Noah recently sent out a suggestion [4] that the ETF (or similar) create 
     an encoding type to represent XML documents for the purpose of carrying 
     XML documents inside SOAP messages. The chair considers this proposal is 
     OK with regard our charter, however its potential impact on our schedule 
     is a major concern. The chair proposes choosing between one of the 
     following options:
       A. discard the proposal
       B. start a list of "next SOAP version" items, and add this proposal 
          to that list
       B.1 add "SOAP with Attachments" to that list
       C. task a couple of people to size the likely time/person cost of 
          pursuing the proposal
       D. pursue the proposal

     - David: There has been some discussion, e.g. What is the relationship with 
              A major concern is the impact on schedule
              WG: how should we proceed?
     - Noah: Friendly ammendment:
             There is a need for something like this.
             We do not have consensus yet.
             Recommends: postpone it/back-burner it
             Interested by the S+A solution.
             Should we get ourselves together about the fundamental model
             of the SOAP envelope?  Is it just the ENV or can it include
             other (external) things (like attachments).
     - David: So you are suggesting another option ("E") which is to start a discussion thread about the nature of a SOAP message and it's relation to stuff outside it.
     - Henrik: We had this discussion a long time ago.
               We decided this is an XML protocol and we're not going to
               deal with other things like binary data, they should be
               solved for XML in general not just for SOAP.
     - David: It seems like there is enough meat for a discussion.  Should
              we even consider discussing it?
     - Noah: It is OK to say S+A is a solution, but then it's not a SOAP solution.
     - Paul Cotton: What we're really talking about is embedding an XML doc
                    inside an XML doc - and this might not be the right
                    WG to solve this problem.
     - David Fallside: Perhaps this goes to some other W3C (XML) WG issue list?
     - David Orchard: XML Packaging WG was started but never took off. Maybe there?
     - Hugo: Several attempts to have packaging mailing list/WG's - but not
             enough real interest.  Next step should be to signal to the
             XML coordination group that we had the discussion and realized
             it's not our role.
     - David Orchard: Now we have clear requirements.
     - David: Sounds like suggestion is postpone and put issue someplace else.
              B - postpone and put on some other XML WGs mailing list, 
              E - learn from the mailing list discussion
              Is the WG ready to decide?
     - Noah: let's vote
     - Asir: This problem is an XML doc in an application defined structure.
              Please clarify.
     - Noah: Document management scenario.
             Just want to be able to put an XML doc in a header.
     - David Fallside: Are we OK with the list of options?
     - David Orchard: ok
     - Hugo: ok
     - David: straw poll - non-binding
              A: 0
              B: 21
              C: 0
              D: 0
              E: 2
              Chair judges that the will of the WG is "B". In that case we need someone to start a thread to discuss where to put the proposal.
     - Noah: Will do it.
     - Paul Cotton: Will help on first draft.


7. Issue 134, xml:base support. Postponed.

11. Misc issues

     The chair proposes that the WG accept the resolutions to the following issues together
      -- Issue 14, see proposed response text
           - David: assume the text is amended per Dug's suggestion (in email).
      -- Issue 112, encoding faultcode.
      -- Issue 53, extension mechanism.
      David: WG - Asks whether WG is OK with considering all 3 issues at once? (WG -- Yes)
      Proposal is accepted without any comment or dissent.
      David: Gudge, please send the issue 14 text?

10. Issue 139, definition of "SOAP Application"

      - David: Do we accept the text in http://lists.w3.org/Archives/Public/xml-dist-app/2001Sep/0139.html
      - WG: yes
      - David: Text accepted. Editors - add it to the glossary, close issue 139.
      - Chris will send resolution text to xmlp-comments


8. Issue 140, identification of self as anon/default actor

     - David: Large amount of email/list traffic on this. What's the current state of this and what's the summary of the discussion?
     - Stuart: In fact, a new thread was started with same Subject line.
               Issue 140: SOAP spec has nothing to say about how an actor determines he is the default actor. Some resolution text has been proposed.
               Jacek liked the wording
               Chris F: +1'd it.
               Noah offered new wording - which Stuart is ok with it.
     - David: What are the final set of words?
     - Stuart: http://lists.w3.org/Archives/Public/xml-dist-app/2001Oct/0053.html
               Spec's discussion of actor is in section 2 and 4
     - David: Proposal: sec 2 - include Noah's text?
     - Henrik: Both proposal text (noah's and stuarts) are ok.
               Discussion went into "how" an actor is determined which he has some concern about.
     - Stuart: That is a different issue
     - Noah: Accept text, table other discussion("how") until later.
     - David: Accept text?
     - Dug: as it is in the note?
     - David: yes - WG vote on text for sec 2?
     - Henrik: Has some editorial question - perhaps?
               scratch "possibly empty"
     - Noah: ok
     - David: Propose the email text without the "possibly empty" phrase.
     WG accepts proposalwithout comment or dissent 
     - David: Directs editors to make the change to the spec text. Asks whether there should be a new issue?
     - Noah: yes
     Hugo will add a new issue
     Stuart will respond to himself (and to "xmlp-comments") with a resolution on 140