Meeting minutes, XMLP telcon, 14 May 2003 1. Roll call Present 16/12 AT&T, Mark Jones BEA Systems, Mark Nottingham Canon, Herve Ruellan Ericsson, Nilo Mitra (scribe) IBM, Noah Mendelsohn IBM, John Ibbotson IBM, David Fallside (chair) Microsoft Corporation, Martin Gudgin Oracle, Anish Karmarkar SAP AG, Volker Wiechers SeeBeyond, Pete Wenzel Sun Microsystems, Marc Hadley Sun Microsystems, Tony Graham Systinet (IDOOX), Jacek Kopecky W3C, Yves Lafon W3C, Carine Bournez Excused AT&T, Michah Lerner BEA Systems, David Orchard Canon, Jean-Jacques Moreau Microsoft Corporation, Jeff Schlimmer Oracle, Jeff Mischkinsky SAP AG, Gerd Hoelzing Systinet (IDOOX), Miroslav Simek Regrets Macromedia, Glen Daniels DaimlerChrysler R. & Tech, Mario Jeckle Fujitsu Limited, Kazunori Iwasa Fujitsu Limited, Masahiko Narita Matsushita Electric, Ryuji Inoue Software AG, Michael Champion Absent DaimlerChrysler R. & Tech, Andreas Riegg IONA Technologies, Eric Newcomer IONA Technologies, Oisin Hurley Software AG, Dietmar Gaertner Sonic Software (Progress), David Chappell Tibco, Don Mullen Unisys, Lynne Thompson Unisys, Nick Smilonich 2. Agenda Review David added one agenda item: Those who have not submitted complete IPR statements to do so. David will send individual emails. 3. Last week's minutes accepted (no objections), with one change: Colleen Evans was present. 4. Action items review (definitive statements recorded on IRC): - publish requirements and usage scenarios as W3C Notes (done) - xml subset thoughts submitted (done) - soap1.2 with wsdl1.1 q/a pending - xkms response drafted and sent by JohnI. The chair of that group has ack'ed it, and agreed to include soap 1.2 as the normative reference - ietf reaction pending - noah's action item still pending 5. Status reports: - Planning of the next F2F: David: got back 2 responses, and checking W3C calendar, he proposes one of two dates - week of June 23 and July 21. The first date coincides with WS-I.org technical plenary. For the week of July 21st, east coast US is preferred. David: July 24, 25 seems like the most promising dates. David: Need to find someone to host that on the East Coast. - XML subsetting David: responses fall into two types - glass half full and half empty. He hasn't seen anything in the emails that bridge those two positions. David: Another way of dealing with it is to take no position on it if we have no consensus. Mark Jones: What interop problems are there by subsetting? What is our role here? David: We have been asked by XML core to provide our opinions on subsetting. Mark: Does not feel comfortable doing this. Noah: There's nothing here about us taking on anything. TAG thought this needs to be studied, and we could help XML core here. David: Mark J, how would you like to respond to core and schema? MarkJ: We haven't achieved a consensus on what constitutes a subset. Not comfortable with XMLP taking a position. Tell them that there are varying opinions within our group. Noah: Friendly amendment. Not only we don't have an opinion, but also ...<<missed the rest>> John: I am in favour of this kind of response. David: Generate text along these lines and send it on to core and schema. Pete Wenzel to generate text to core and schema. To be done by end of the week. Post to member list, and if no major comments, David will send it on. -Q/A for SOAp 1.2 in WSDL 1.1 David: It would be good to know the outcome of the discussion between WSD WG and soapbuilders. We'll get a response from the WSD WG next week. 6. Media registration MarkN: No discussion on the media types list. The person who raised the question has accepted that our SOAP 1.2 media type is to be used. But this issue is more general than XMLP. MarkN: We could make explicit in our draft to state that it is only for SOAP 1.2. David: We should confirm that he is satisfied with the modified draft. Mark Baker has the editor's copy. ACTION: MarkN will make more explicit in the IETF draft that the media type is applicable to SOAP 1.2, and also to check with John Kemp that he is satisfied. Then resubmit the draft. 7. PR Issues David: Some left over from the transition, and some fom the larger public. We will receive comments from ACs. Some comments from AC reps may be anonymised. Technical issues will show up on our PR issues list. The bar for changing anything in the PR is very high. David: We will be updating the editor's copy. We will submit that to the director. The hope is that what comes out is what we put in, but the director can remove some changes before making it a final recommendation. Need to record every single change made. Jacek: Will the director consult us if he wants to remove some changes? David: Yes, he will. Yves: there will be consultation between the director and the WG. Issue 423: Anish: Test T20 refers to an older version of the spec. T20 should be removed. Gudge: Could the test check that it accepts both qualified and unqualified message? Anish: If we removed T20, there'd be a gap. David: Is T20 referred to anywhere else? Anish: Yes, it is used for an assertion about faults. I am neutral. Gudge's proposal is OK, too. Jacek: The test should be removed. Agreed (no objections) to resolve Issue 423 by removing T20 and all its references. Anish will make the change. Anish will send email to originator and xmlp-comments. Issue 424: Anish: Do not recommend making any change to T77. Resolved (no objections) to retain T77 as is. Anish will do the appropriate response. Issue 425: David: Noah argued against Elliot's pov in a message. Gudge: This is not a change since CR. The language is more explicit now. What do we do about comments like this? Noah: Do not want to imply that you are probably right but it is too late. This is not the case where we have made a mistake. David: Elliot notes that it was there in CR but now it is more explicit. Gudge: Noah is saying that this is not apporpriate. Noah: There is consensus in the WG that this is not a serious enough issue, and would not be one at any stage. At most, this is a minor shortcoming. David: For a clear resolution, we need to spend time to craft the text. Gudge will generate some text for review by the WG. Issue 425 will be closed after we get Gudge's text. Issue 426: Anish: Why do we have text/html in the first place? Nilo gave some explanation. He suggest that we accepts the second proposal. Noah: From a client's pov, you can accept one of several. Jacek: Since we added the use of HTTP GET, it is important that we should how SOAp interacts with the Web. Anish: I'm reading this out of context. Resolved (without objection) by accepting Eliott's second suggestion to add a q value to show the preference between the different media types. Nilo will send mail to Eliott and xmlp-comments. Issue 427: There was a long email thread, the originator accepts that the example is OK. Volker: All the examples in the whole document are not encoded in UTF-8 . If you convert your browser to UTF-8 you will not see the name in the example, but some strange characters. Jacek: This is not a problem. The display of the whole primer document....<<missed the rest>> Gudge: The first problem is whether or not there needed to be an encoding declaration. The second problem is how it looks in its html and xml renditions in different browsers. Volker and Gudge examined the source. Gudge and Volker will work on this offline. Gudge: The text of the document (Primer) won't change. There may be some change to the underlying "code". 8. Attachments David: Have we identified any issues? Jacek: One of the issues is where does inclusion occur - in the binding, or the processing model, or the header? Noah: Is PASWA in track with the SOAP processing model? Jacek: While PASWA is not in SOAP 1.2 terms, the PASWA document takes one side but there are other proposals. Noah: Have we accepted that bindings cannot introduce headers? <<missed much more that he said>> Jacek: Binding can introduce headers, but not sure if they can "eat" headers on the other side. Gudge: This is a question we need to answer, but it is not the first question we need to answer. There are a whole number of questions we could answer without aanswering this one. Noah asked for a clarification from Gudge. <<long technical discussion on clarifications between Noah and Gudge which I could not capture>> David: We should capture this idea when xbinc is visible and when it isn't, and do it in a rewrite of the proposal. MarkJ: Are there any other issues? Marc: Capture all high level issues. Jacek has identified two issues which he will send to dist-app. Meeting adjourned.