XMLP WG Telcon minutes, 12 May 2004

>1. Roll call. Scribe for minutes selected from attached list. Actions to be
recorded on IRC. (830PT + 5)

Present, 12/10
BEA Systems, Mark Nottingham
Canon, Herve Ruellan
IBM, John Ibbotson
IBM, David Fallside
IONA Technologies, Suresh Kodichath
Microsoft Corporation, Martin Gudgin
Nokia, Michael Mahan
Oracle, Anish Karmarkar (scribe)
SAP AG, Gerd Hoelzing
SAP AG, Volker Wiechers
SeeBeyond, Pete Wenzel
Sun Microsystems, Tony Graham

BEA Systems, David Orchard
Canon, Jean-Jacques Moreau
IBM, Noah Mendelsohn
Microsoft Corporation, Jeff Schlimmer
Oracle, Jeff Mischkinsky
Sun Microsystems, Marc Hadley

W3C, Yves Lafon

>2. Agenda review, and AOB (835 + 5)
>The Important Stuff
>We need to decide on handling XML 1.0/1.1 in SOAP 1.2, resolving issue 443,
and comments on the LC specs. Once completed, the Chair will ask whether we can
enter a final spec review so that we can request Last Call.

DavidF: Plan to go to LC around 24th May. Then four weeks of LC.
DavidF: need to have a F2F, one option is to do it after LC, or after CR (to 
  get ready for PR). Depends on the number of comments for LC. CR target date,
  assuming that we come out of LC about 21 of june, is 12th of July. Then 4
  weeks of CR.
Gudge: depends on how many comments we get for LC. I think we will probably get
  more comments during CR, might make sense to have a F2F after CR.
DavidF: Lets us think about this.


>3. Approval of minutes, 5 May and 28 April will be posted to the mailing lists
shortly (840 + 5)

No objections to approving 5th may and 28th april minutes.
Both of them approved.

>4. Review action items, see http://www.w3.org/2000/xp/Group/Admin/#pending.
These action items are taken from the XMLP member page. Some AIs may be discussed
under other agenda items (845 + 5)
>2004/03/14: Michael
>Update the xbinc:include FAQ and turn it into a WD by april 21
>PARTIAL, ed copy at http://www.w3.org/2000/xp/Group/4/05/FAQXOPInclude.html,
>2004/04/21: Mark
>Submit application/soap+xml to IETF
>2004/04/28: DavidF
>Email to Martin Duerst re. old/new procs for 2 media registrations
>DONE, see http://lists.w3.org/Archives/Member/w3c-archive/2004May/0041.html
>2004/04/28: Mark
>Send app/soap_xop+xml txt to eds for incorporation into MTOM


>2004/04/28: Editors
>Incorporate http://lists.w3.org/Archives/Public/xml-dist-app/2004Mar/0016.html
in XOP (resolution of 464)

Herve: think it was done last week.
Herve: yes, it is done.

>2004/04/28: DavidF
>Note in MTOM LC cover material that it has supeceded Att Feature doc
>2004/04/28: DavidF
>Initiate 'parking' of Att Feature doc at MTOM LC
>2004/05/05: DavidF
>Ensure entire WG knows there is a 1.0/1.1 decision scheduled for may 12
>DONE, http://lists.w3.org/Archives/Member/w3c-xml-protocol-wg/2004May/0016.html
>2004/05/05: Herve
>Draft XOP prose to make schema non-normative and incorporate all schema
constraints into the prose. Due COB Europe time Friday May 7
>DONE, see http://www.w3.org/2000/xp/Group/3/06/Attachments/XOP.html
>2004/05/05: MTOM editors
>Starting 5/12, reflect SOAP 1.2 XML 1.1 decision (infoset?) in MTOM. May be
only editorial cleanup.
>2004/05/05: Noah
>Track down lexical form rules for xsd:anyURI. In particular, is it restricted
to XML 1.0 characters?

DavidF: is this critical?
Gudge: Let me dig a bit on this.
Done - by Gudge.

>2004/05/05: Anish
>(a) as editor of media type doc, check that media type attribute text in XOP is
the same as corresponding text in media type doc (b) propose alternate text for
XOP to appropriately reference media type by COB May 7

Anish: still pending.

>5. Status reports and misc (8.50 + 10)
>-- (Pending items)
>-- Representation Header document, status update

No update.

>-- Media types registrations, i.e. "application/soap+xml" and "application/
soap_xop+xml" (MarkN)

Mark not here, no report.

>-- FAQ regarding MTOM/XOP instead of XInclude, see http://www.w3.org/2000/xp/

DavidF: thought it was going to be a WG Note.
Mike: took the template (based on instructions from Yves). Should it be a WG
DavidF: yes, we should just publish a WG Note. I would like to see this FAQ 
  have some more weight, and so it should be a WG Note rather than plaintext.
Mike: there is more to it then fixing up the header section with the template.
  There is a process issue.
DavidF: There isn't much process involved here. Lets get that done by next week.
Mike: I see a couple of typos and logic issues, that needs to be fixed.
DavidF: please make editorial adjustments as necessary.

>-- XMLP/WSD Task Force and WSDL Media Type document, currently at http://
lists.w3.org/Archives/Public/public-ws-media-types/2004Jan/0001.html (Anish)

Anish: new draft checked in. There will be a Ed note to point to issue 443. The
  WSD WG may put this agenda during the F2F next week.

>-- Volunteer sought to review Web Services I18N docs, http://www.w3.org/
International/ws/ws-i18n-scenarios-edit/Overview.html and http://www.w3.org/TR/

DavidF: they asked if we could take a look at the material. Two docs -
  scenarios and requirements. I would like to get two volunteers to do this.
  The mission is to figure out whether there is anything pertinent to SOAP and
  any other thoughts on the global issues about I18N and Web services.
TonyG: will volunteer for the requirements doc.
DavidF: let us have a report back on 26th of may.

>6. Attachments (9.00 + 60)
>-- Placeholder for pending items
>o Update on Copyright and IP statements
>-- Issue discussion.
>o Rec20, http://www.w3.org/2000/xp/Group/xmlp-rec-issues.html#x20, SOAP 1.2
and XML1.1. For summary of issues see http://lists.w3.org/Archives/Public/
xml-dist-app/2004Mar/0024.html and http://lists.w3.org/Archives/Public/
>We will decide this week between two proposals for handling XML 1.0 and 1.1,
see http://lists.w3.org/Archives/Member/w3c-xml-protocol-wg/2004May/0016.html
and Yve's clarification of proposal (2) at http://lists.w3.org/Archives/Member/
>With this decision in hand, we should revisit our prior decisions on the same
topic and their impacts on our docs. Specifically, our agreement to:
>i) at the infoset level we will accept any version of XML and in our spec we
will note that whereas the infoset may support any version of XML, individual 
node may be more restrictive.
>ii) at the HTTP binding level we will restrict application/xml+soap to XML 1.0 
and in our spec we will note that other media types may not impose such restrictions.
>iii) to accept Noah's errata text for SOAP 1.2 Part 1 Rec but to not specify a 
standard fault, see http://lists.w3.org/Archives/Public/xml-dist-app/2004Apr/
0012.html. However, see additional commentary at http://lists.w3.org/Archives/
>iv) text regarding XML 1.0/1.1 from Herve to apply to MTOM: See proposal at 
>And the items on which we had not yet reached agreement:
>v) text regarding XML 1.0/1.1 from MikeM to apply to SOAP 1.2 Part 2 Rec: See 
proposal at http://lists.w3.org/Archives/Public/xml-dist-app/2004Apr/0027.html 
and thread http://lists.w3.org/Archives/Public/xml-dist-app/2004May/0004.html
>vi) clarifying text for part1 binding framework to clarify serialization of 
the infoset from Yves: See proposal at http://lists.w3.org/Archives/Public/

DavidF: is there anyone who has not read the two proposals?
No reponse from any one.

Gudge: question about proposal 2 -- it says "allow 1.1 envelope ..." does this
  mean at the infoset level or for the HTTP binding. From last week's minutes it
  seems like it is about the serialization and not the infoset.
DavidF: that is what I recall too.

  Proposal 1:
    Favor: Gudge, Tony, John, Suresh, Pete, Herve, Gerd, Anish, Mike, Noah, 
    Lie-down-against-this: none

  Proposal 2:
    Favor: none
    Lie-down-against-this: Gudge

DavidF: based on Yves' email I am going to assume that Yves is "lie-down-against" 
  proposal 1
DavidF: we know that Yves is for proposal 2. We should also note that MarkN
  voted for proposal 1 and against proposal 2 (but no lie down in the road).
DavidF: chair rules/notes that proposal 1 is the will of the WG. Also notes 
  the W3C objection to proposal 1.

DavidF: as i said in my email we also thought about leaving tracks in the rec
  about the fact that we might reconsider this if schema WG were to come up
  with a solution for XML 1.1. This might go in the errata or alerting the TAG
  or any other things. Opinion?
MarkN: not particularly. What are the implications?
DavidF: SOAP env are xml 1.0 at the infoset level. HTTP binding will be xml
  1.0. But other serialzations many do xml 1.1
MarkN: Will soap env not contain new 1.1 char in QNames?
DavidF: the motivation for that is that we have no way of defining this in
MarkN: we should then bring it up to the TAG or the XML coordination group.
DavidF: can you put together an email to capture this? In time for next week?
MarkN: yes

DavidF: we need to go back again to the rec and figure out what we are going to
  change to reflect our decision. Although, I am not sure if we added anything 
  to the errata, we only added text to MTOM/XOP. (i) and (ii), (in the agenda 
  above), were just decision.  (iii) was proposed text -- we did not change 
  any text in errata. Do we need to do anything with it?
Gudge: don't think so.
DavidF: do we want to say anything about binding-dependent fault if the XML
  version is not supported?

General agreement that we probably should say something about faults and XML
versions (for section 4.2).

For section 5 -
Gudge: the infoset may be serialized as anything, but the infoset itself is
  1.0. That is the proposal that we just accepted. If we are only dealing with
  the NEL char then in XML 1.1 the serializer could take 'D' or 'A' and convert
  it to NEL. The other side would do the reverse.
DavidF: so looks like we don't need (iii) and say that bindings may accept
  other. We will need that stmt, but we don't need Noah's errata text.
Gudge: yes.
DavidF: we need someone to propose text for errata by next week.
Gudge: I will do it.

Anish: we did agree to say something about faults and XML version at the
  binding level.
DavidF: yes, I think Noah's text provide this for the binding level fault.
  We already decided that we don't need a standard fault. So we need a volunteer
  to do some ed work on Noah's text.

David: Herve is any text from (iv) relevant?
Herve: i think I will have to look at it and make a new proposal.
DavidF: by end of week?
Herve: yes.

David: Mike you have come up with the proposal in (v). How does our decision
  affect this?
Mike: two changes, the 1st one does not apply, the 2nd change only partially
  applies. But it is more ed. and not necessary -- the 'MAY' part.
Anish: I think the 'MAY' part is still applicable
Mike: if we replace 'representable' with 'serializable' then perhaps.
DavidF: Mike, can you take a look at the text and make a new proposal in light
  of our decision.
Mike: sure.
Mike: Noah made some changes to my text, I will roll that in.

DavidF: another piece of text that was proposed by Yves -- (vi) above.
MarkN: is he talking about infoset or serialization? if infoset then we support
  only 1.0. But I thought he was talking about the serialization. The problem is
  that the range of char that an infoset contains, in many people's mind,
  conflated with serialization.

WG discusses (vi)

Mark: if you edit this to include "... serialization, example XML 1.0" this 
  might make it very clear + some more edits.
Mike: in the addition, the 'must' should be 'MUST'.
DavidF: hard to do wordsmithing on the phone. I will aske Yves to reevaluate
  his proposal in light of our decision.

DavidF: we still cannot close rec20, we need to take care of (i)-(vi). I don't
  think we can go to LC without this issue closed, even though it is a rec issue.

Gudge: xsd:anyURI issue -- URI according to 2396 can contain only ASCII char 
  only with escaping mechanism. 1.1 will allow you to write down URI that you 
  could not write down with 1.0. But anything that you write in 1.1 you can 
  write in 1.0.

Discussion on this issue

DavidF: Gudge could you document this and send it to the ML.
Gudge: yes

>o Rec22, http://www.w3.org/2000/xp/Group/xmlp-rec-issues.html#x22, Allowed 
characters in infoset for XML serialization. We decided at the 21 April telcon 
that Rec22 would be covered by our resolution to Rec20.

DavidF: 22 is covered by 20.

>o 443, http://www.w3.org/2000/xp/Group/xmlp-issues.html#x443, Media-type 
information for binary data. We agreed in principle at last week's telcon that 
the Media type doc satisfies resolution of this issue (see http://lists.w3.org/
Archives/Member/w3c-xml-protocol-wg/2004May/0007.html). We agreed that Anish 
would perform a final check on the doc's text (see Anish's AI above) before 
closing the issue.

DavidF: we need to make that sanity check on the doc's text.
Anish: i will do that.

>Comments on specs:
>-- MarkN on XOP, http://lists.w3.org/Archives/Member/w3c-xml-protocol-wg/
2004Apr/0051.html, and responses from Herve http://lists.w3.org/Archives/Member/

DavidF: we had comments from Mark and reponse from Herve. Where are we at?
Mark: no strong disagreement. Fine with what Herve came up with.
DavidF: Herve can you put them in the spec but flag them. We need to have a
  discussion about the comments on the doc. And we will have to ask if people
Herve: ok. will flag the big stuff and not flag the ed stuff.
DavidF: please send an email to the WG regarding this when you are done. Can
  you also work with Yves and myself to put this in LC format. Next week I will
  ask the WG if we are ready to go to LC.

>-- Any other comments? Are we ready to enter final review before Last Call?

Anish: Noah has sent comments on the Representatin Header doc. I +1'ed it.
  Since Yves is the one who is sort-of responsible for the sections (on which
  comments were sent), I'm waiting for Yves to respond.

Mark: Would like to do informal interop of XOP/MTOM, so if anyone is interested
should let me know.

>7. SOAP 1.2 Recommendation maintenance (postponed)

End of Call