Web Services Description WG call
30 Sep 2004

<scribe> Scribe: hugo

Approval of minutes

Sept 23 minutes at http://lists.w3.org/Archives/Public/www-ws-desc/2004Sep/0043.html approved

Review of Action items & Editorial actions

Jonathan asks editors to review their list of action items at http://www.w3.org/2002/ws/desc/4/lc-issues/actions.html and let him know when one has been done

Chair goes throught his list (see agenda)

--------------------------------------------------------------------
3.  Review of Action items [.1].  Editorial actions [.2].

PENDING   2004-04-01: Marsh will get schema tf going.
PENDING   2004-08-04: Jonathan to check the policy with AB 
                      and team and perhaps set 
                      up a ML for testing.
PENDING   2004-09-02: Bijan to create stylesheet to generate a
                      table of components and properties.
PENDING   2004-09-09: Roberto to classify our errors as fatal or 
                      non fatal. (LC5f)
DONE [.8] 2004-09-14: Arthur to submit proposal on conformance 
                      requirements for XML serialization (of 
                      WSDL and schema documents) 
DONE [.3] 2004-09-14: JMarsh to take issue LC6d to XML Core WG 
DONE [.4] 2004-09-15: Arthur to (re)write a subsection of the 
                      spec to show exactly how it looks with 
                      the Zed notation included before next 
                      telecon.
DONE      2004-09-15: Hugo to investigate potential options for 
                      get the Zed version published in W3C 
                      web site.
DONE [.5] 2004-09-15: Tom and Anish to figure out the right 
                      words for a more meaningful overview/
                      introduction to the problem we're 
                      trying to solve.
PENDING   2004-09-15: Jonathan to ask XForms folks to review 
                      WSDL.
PENDING   2004-09-16: Editors to move App C to RDF Mapping spec, 
                      except the frag-id which will move 
                      within media-type reg appendix.
PENDING   2004-09-16: Editors to fix paragraph 6-9 of section 
                      2.1.1 moved into 2.1.2
                      which talks about the syntax.
PENDING   2004-09-16: Hugo to get a URI to use for DTD example 
                      in Appendix E.1 (LC38)
PENDING   2004-09-16: Glen to CC Asir on mail to Marc re: SOAP modules 
                      and features (LC18, LC29b)
PENDING   2004-09-23: DBooth to get namespace and URL for 
                      Arthur to use for Z notation.
DONE [.9] 2004-09-23: DaveO to add more descriptive text 
                      regarding LC29d.
DONE [.2] 2004-09-23: Marsh to check we have enough Ed AIs.
DONE [.6] 2004-09-23: Hugo to investigate how to resolve LC 45.
NEW       2004-09-30: Marsh to ask Glen about how LC9 is going.

[.1] http://www.w3.org/2002/ws/desc/#actions
[.2] http://www.w3.org/2002/ws/desc/4/lc-issues/actions.html
[.3]
http://lists.w3.org/Archives/Public/public-xml-core-wg/2004Sep/0078.html
[.4] http://lists.w3.org/Archives/Public/www-ws-desc/2004Sep/0041.html
[.5] http://lists.w3.org/Archives/Public/www-ws-desc/2004Sep/0048.html
[.6] http://lists.w3.org/Archives/Public/www-ws-desc/2004Sep/0049.html
[.7] http://lists.w3.org/Archives/Public/www-ws-desc/2004Sep/0035.html
[.8] http://lists.w3.org/Archives/Public/www-ws-desc/2004Sep/0060.html
[.9] http://lists.w3.org/Archives/Public/www-ws-desc/2004Sep/0061.html

Hugo: regarding the format required for the publication of the Z notation document, the normative version must be the one with unicode chars
... we can provide an alternate one with graphics instead of them
... tests on lists.w3.org with Paul have shown that it works for lots of people

Administrivia: November 9-11 (Sunnyvale, CA) registration

No questions

Last Call Issues

Jonathan: still working on this; I just need to put Kevin's in

Arthur has questions about changes that we've made that aren't a result of a LC issue

Jonathan: so far, we only did that as a result of Bijan's F2F presentation

Media Type Description Last Call status

Anish: all the F2F changes are done

<Marsh> http://lists.w3.org/Archives/Public/www-ws-desc/2004Sep/0048.html

Anish: the only issue left is the proposed text about expanding the introduction text

DBooth: does that affect the media type registration?

Anish: no

Jonathan: any comments?
... we have consensus to accept Anish's proposal

RESOLUTION: http://lists.w3.org/Archives/Public/www-ws-desc/2004Sep/0048.html accepted as new intro text for Media Type Description Note

Jonathan: we have several options for going forward
... maybe it would be best to hear from XMLP before taking our vote to go to LC

<scribe> ACTION: Jonathan to contact XMLP to review the Media Type Description document in order for us to go to LC with it

Anish: I would like people to make sure I did the right thing with my edits

Jonathan: we could decide to go to LC next week if everything goes right

<Marsh> ACTION: Working Group to review Media Type note in preparation for LC vote next week.

<dbooth> http://dev.w3.org/cvsweb/~checkout~/2002/ws/desc/wsdl20/wsdl20.html?content-type=text/html;%20charset=utf-8#ietf-draft,text

Media Type Registration

DBooth: I have a related item about the media type registration
... there is an editors note at the top of appendix A saying that the WG does not agree on the media type

Jonathan: in the XML Core WG, somebody thought that we did not need to register a new media type
... however, I heard that there was a W3C policy pushing us to do so

<pauld> observes that application/soap+xml media type was published only this week: http://www.ietf.org/rfc/rfc3902.txt

David: I think that it's an informal policy
... moreover, it seems to me that it's something we would want to do

Tom: why is it application/ instead of text/?

Jonathan: an RFC (not sure which number) prefers application/* to text/* for XML vocabularies, saying that XML's not for human consumption

<pauld> points tom at this article: http://www.xml.com/pub/a/2004/07/21/dive.html

Jonathan: originally, you could serve XML as application/xml or text/xml
... now, they are pushing people to use application/xml

Tom: I am a little worried about interoperability
... it sounds that my browser wouldn't know how to deal with a WSDL document with such a media type

Arthur: I believe that it's an IETF RFC which says that we should register a media type, and use application/*
... and we could specify what to do in cases where the document is read by a Web browser

<dbooth> RFC 2048 discusses media type registration

Jonathan: there might be an issue here about registering a new media type, interacting with existing tools, etc.

<dbooth> RFC 3023 discusses registration specifically of XML media types

[ people do some tests with browsers ]

Arthur: the purpose of the +xml convention is that you can deal with such cases gracefully

<dbooth> http://www.w3.org/TR/webarch/#xml-media-types

DBooth: the Web Arch says: In general, a representation provider SHOULD NOT assign Internet media types beginning with "text/" to XML representations. -- http://www.w3.org/TR/webarch/#xml-media-types

<dbooth> [[

<dbooth> In general, a representation provider SHOULD NOT assign Internet media types beginning with "text/" to XML representations.

<dbooth> ]]

Tom: I think that a new media type is going to be trouble
... Axis doesn't pay attention to the media type of WSDL documents

<sanjiva> If we don't introduce a mime type don't we lose the fragment stuff?

Jonathan: do we want to track this as an issue?

<sanjiva> If that was discussed I'll shut up

<dorchard> We have to do media type registration for frag-id.

Arthur: we can do some tests

Jonathan: why don't David or Hugo set up a test resoure?

<dorchard> what is the issue?

<Marsh> Whether to recommend people serve up WSDL as application/wsdl+xml

<Marsh> Or whether application/xml is a better choice.

<scribe> ACTION: Hugo to set up a application/wsdl+xml on the W3C site for tests

<dorchard> it has to be wsdl+xml

Hugo: but WSDL documents are targetted to WSDL processors, not people through Web browsers

<Zakim> dbooth, you wanted to ask what impact the outcome would have

Tom: I often look at some

DBooth: what would the outcome of the test be?
... it will not be a WSDL-specific problem

<pauld> i'd create a perl CGI first line print "Content-type: application/wsdl+xml\\n"

Jonathan: agreed, it would be a problem for the TAG

DBooth: will that affect our registration?
... I discovered that we were supposed to do this a while ago

Jonathan: we should just go ahead for now

DBooth: with application/wsdl+xml?

Jonathan: yes

DBooth: the deadline for us was 2 weeks before LC
... our LC ends on Monday
... why don't we extend our LC period for them?

Jonathan: what about the 18th?
... schema and maybe XForms asked for extensions too

<asir> displays fine for me

Decision to use "WSDL" for the Macintosh File Type Code

Zed notation

Jonathan: let's separate the content and display discussions

<Arthur> http://lists.w3.org/Archives/Public/www-ws-desc/2004Sep/0076.html

Arthur: the advantages is that it's precise and can be validated
... see screenshots at http://lists.w3.org/Archives/Public/www-ws-desc/2004Sep/0076.html

Jonathan: we always expressed interest; does that meet our expectations?

DBooth,Sanjiva: it exceeds my expectation

Amy: the symbols do not talk to me [ paraphrased by scribe ]

Jonathan: we talked about including a short intro to describe common Z notation symbols

Amy: it's better indeed, even though there still is a risk of making this dense for people not used to it

Roberto: it looks great; there is duplication between English and Z, which is troubling me as an editor
... it's twice as much work
... and we don't know what the reaction of our readers will be; we wouldn't want to have people *not* read the WSDL 2.0 because of this

Kevin: I'm concerned that we're adding another layer to the spec; people have complained that there are too many layers with the abstract components, the mapping to XML, ...

Arthur: we previously discussed having a reader's digest version

<dbooth> Kevin, I think "normal readers" can primarily read the Primer. The spec is more intended for people who need the precision.

Arthur: there could be another version, lighter one, with Z
... however, we wouldn't want another document for that; we could have English statements very close in writing as well in placement to the Z statements

<dbooth> +1 to Arthur's comment. MUCH easier to maintain correspondence if the Z is integrated, rather than in a separate document.

Dave: this seems weird to me to add the Z notation in light of the decision we made about spec simplification

DBooth: I view the Z notation to help me to clarify, whereas I don't see the boilerplate stuff as necessary

Jonathan: it seems that we need to publish a draft with the notation in to see how people are going to react to it

Roberto: that means reformating the text, and potentially having to take it out afterwards; that's a lot of work!
... it would be better not to remove any text for now, just to add Z so that it can be removed easily
... we need feedback to make a decision

Jonathan: we could get feedback from our companies

<Arthur> I can keep the Z separate from the existing text

Jonathan: we could publicize Arthur's example too

Arthur: we can do the Z as a net addittion, in a different namespace, so that it will be easy to deal with later on

Kevin: is Z WSDL-specific?

Arthur: I did it for the Infoset too
... as a test case

Kevin: maybe we could recommend this to other WGs

Jonathan: I hear that we do like this but we're concerned about our users' feedback
... I suggest that we should just include it and we'll see what feedback we get

DBooth: it seems to me that gathering more feedback is the slower route; we've always endorsed this, so we should just go ahead with this

Jonathan: the proposal is to go ahead and include the Z notation in the drafts

Kevin: I would like to have another version, more readable, without the Z notation

<sanjiva> hmm not passing thru

<sanjiva> hang on

Consensus to accept the amended proposal

<Marsh> RESOLVED: Appoint Arthur as Part 1 editor

Arthur is appointed as Part 1 editor to add Z notation

<sanjiva> ok http://203.94.84.117:9998 returns application/wsdl+xml and http://203.94.84.117:9999 returns text/xml

Hugo: we also need Z in Part 3

Arthur: Agreed, we need Z everywhere we make formal statements
... but let's just start with Part 1 for now
... I would like a consistency check be part of the build process

Hugo: Arthur and I need to talk about it

RESOLUTION: Integrate Z notation to drafs

<scribe> ACTION: Arthur to add Z notation to Part 1

Issue LC6d: QA Review on WSDL 2.0 Part 1, Technical comments

Jonathan describes http://lists.w3.org/Archives/Public/public-xml-core-wg/2004Sep/0079.html

Jonathan: XML Core suggested a naming prefix (not a namespace prefix) to hint that our fragment ids are WSDL-related

Arthur: sounds reasonable

<sanjiva> Jonathan: Can we decide the application/xml thing? we have the evidence we need.

Arthur: we could use "wsdl-"

Jonathan: any objection to "wsdl-"?

Arthur: actually, I like "wsdl." better

Jonathan: any objection to "wsdl."?

RESOLUTION: we have no issue about our scheme name
... we will preface our names with "wsdl."
... LC6d closed

<scribe> ACTION: Editors to add "wsdl." to XPointer syntax

Media type (continued)

Sanjiva: we have evidence that application/wsdl+xml does not display well in lots of popular browsers

People report that application/wsdl+xml renders under IE

<dmoberg> both 9999 and 9998 in my version of IE is displayed

<sanjiva> +1 for Hugo's proposal

<Arthur> I opened http://www.w3.org/2002/ws/desc/2/04/Service1.wsdl with IE and it first opened in Wordpad because I had made that file association.

<Arthur> I changed the .wsdl files association to IE and http://www.w3.org/2002/ws/desc/2/04/Service1.wsdl opened as an XML document.

Hugo: I think that we need to register application/wsdl+xml; the rendering problem is orthogonal to me, and that Tom may want to ask the TAG want they think about media types not rendering about browsers not rendering ...+xml documents well

DaveO: we define frag ids, we need to indeed

<Arthur> This works because of the .wsdl extension.

DaveO: we could report our experience about the deployment of our application/...+xml media type

<Arthur> IE downloaded it and then matched the .wsdl extension, not the Mime Content Type

Jonathan: maybe we should stay silent about what the right thing to do is

Tom: maybe we should say nothing and let the problem work itself out

Arthur: actually, I originally proposed text/xml but got shot down

<pauld> wants to register application/wsdl+xml and let the world sort this out ..

Jonathan: proposes to do nothing about it, and maybe do something about the rendering issue later

Arthur: we can also post a bug report for Firefox

<dbooth> http://www.ietf.org/rfc/rfc3023.txt

[ Discussions around the +xml syntax ]

<dbooth> [[

<dbooth> This document standardizes five new media types -- text/xml,

<dbooth> application/xml, text/xml-external-parsed-entity, application/xml-

<dbooth> external-parsed-entity, and application/xml-dtd -- for use in

<dbooth> exchanging network entities that are related to the Extensible Markup

<dbooth> Language (XML). This document also standardizes a convention (using

<dbooth> the suffix '+xml') for naming media types outside of these five types

<dbooth> when those media types represent XML MIME (Multipurpose Internet Mail

<dbooth> Extensions) entities.

<dbooth> ]]

<dbooth> -- from http://www.ietf.org/rfc/rfc3023.txt

RESOLUTION: no change with regards to our media type