W3C Forms teleconference December 1, 2010

* Present

Mark Birbeck, Web Backplane [IRC]
Philip Fennell, MarkLogic
Steven Pemberton, CWI/W3C (chair)
Leigh Klotz, Xerox (minutes)
Uli Lissé, DreamLabs
Kurt Cagle, National Archives [joined late]
Nick van den Bleeken, Inventive Designers [joined late]
Alain Couthures, AgenceXML [joined late]

* Agenda


* Meeting

Steven Pemberton: We have two potential regrets from Erik Bruchez and Nick van den Bleeken, and John Boyer hasn't showed.

* Action Item Tracker

Steven Pemberton: We may be able to initialize the table for the bug tracker. They don't like us re-initializing it; although we have 300 issues in the system, none of hte URLs have any meaning as we've never used them.
Leigh Klotz: They're unusable; is there a value for that?
Steven Pemberton: I suppose we could just close them all but frankly I'd rather just start fresh since we've never really used it.

* Status Report to HCG (copy)


Steven Pemberton: I send these in bi-weekly. I'll be sending another today.

* XForms 1.1 load erratum

http://lists.w3.org/Archives/Public/www-forms-editor/2010Nov/0000.html http://lists.w3.org/Archives/Public/www-forms-editor/2010Nov/0001.html

* New Invited Experts

Steven Pemberton: We're expecting invited expert Kurt Cagle today. Dan MacCreary is also an invited expert
Kurt Cagle: [joins] I'm working as a consultant for Lockheed on the US National Archives project and have been working with XForms for a long time. I'm doing the data architecture for the national electronic archives project. I'm working with XQuery and XForms to develop pieces as part of. I'm not the XForms developer, but I'm trying to make it easier for the XForms developers on staff.
Steven Pemberton: That's fantastic news.
Steven Pemberton: Let's all introduce ourselves to Kurt.

Steven Pemberton: We're using the wiki to coordinate distributed editing of specs.

* Action Item Tracker [again]

Nick van: [joins] I will look at the script.
Steven Pemberton: And I'll send it to W3C staff to run it.

* XHTML + XForms Validator

http://lists.w3.org/Archives/Public/public-forms/2010Nov/0025.html http://lists.w3.org/Archives/Public/public-forms/2010Nov/0026.html Second message has link to XHTML+XForms validator from relaxed.vse.cz

Leigh Klotz: It uses NVDL. I can't get it to work?
Steven Pemberton: What's our current status?
Leigh Klotz: How do we add to XHTML elements using modularization?
Steven Pemberton: You can.
Leigh Klotz: And we should talk to Shane McCarron about RNG integration.
Steven Pemberton: Yes, we're about to publish the Relax version as a note.
Leigh Klotz: Is there an advantage to using the XSD version because it's a Recommendation?
Steven Pemberton: No.
Leigh Klotz: What are the mechanics?
Alain Couthures: [joins]
Steven Pemberton: We should talk to Shane McCarron. I'll send a message to Shane, CC'ing you.

Action 2010-12-1.1: Steven Pemberton to contact Shane McCarron to set up meeting to discuss XHTML M12N using RelaxNG for XHTML+XForms integration.

* Nested Bind Question


Steven Pemberton: We answered Claudius Teodorescu's question.

* Input Controls and Datatypes

http://lists.w3.org/Archives/Public/public-forms/2010Nov/0028.html http://lists.w3.org/Archives/Public/public-forms/2010Nov/0029.html http://lists.w3.org/Archives/Public/public-forms/2010Dec/0003.html Question from Alain Couthures

Steven Pemberton: I think this is too big a change for 1.2 but a good path to go down.
Kurt Cagle: It seems like the ability to declare larger objects and the ability to do bindings (XBL2, WebBL) are all related in a general extensions issue that I tend to agree is something we can't get into 1.2. I wonder if a background task to develop a note would be good.
Steven Pemberton: XBL has been bubbling in the background but we've never run with it. Several implementations use it as a basis for doing their controls. I think that it's a valuable way of moving to the future; the problem is that XBL within W3C is a little hazy. We've had the XBL2 spec with no interest outside XForms. Ian Hixson recently made a major uncoordinate change to it, which makes it strictly speaking useless for us. I think there is a future in XBL. We had a discussion a few weeks ago and for example we'd like to see XPath selectors.
Kurt Cagle: I tried that in the SVG side and it didn't work.

Kurt Cagle: Implementations aside, it might be nice to develop in the Wiki something like "import modules" in XQuery. It might be worth creating a page where we can talk about functional modularization for XForms. The function extensions is one area, but if we look at it from a modularization standpoint as with XQuery, it would be useful.
Steven Pemberton: Sounds like an action item.

Action 2010-12-1.2: Kurt Cagle to create wiki page to discuss XForms and E11Y and M12N.

Nick van: [irc] How XBL is used in Orbeon: http://wiki.orbeon.com/forms/doc/developer-guide/xbl-components-guide
Steven Pemberton: Yes, Mark Birbeck did a version of Google Maps using XBL and XForms showing the power of XBL.
Steven Pemberton: Are we done with this topic for now?
Steven Pemberton: Alain, if you want to put notes on the wiki about this, please go ahead.

* XForms Futures discussion on XSLTForms List


Alain Couthures: I proposed two extensions: XQuery and XHTML5 menus. There was interest but also concerns about extensibility with widgets. The concerns were more about XForms than XSLTForms.

Steven Pemberton: There are two issues you stated; the XQUery one.
Leigh Klotz: Follwing the link http://sourceforge.net/mailarchive/forum.php?thread_name=AANLkTin4Cd609NVQxbwnMCmkwzCsX33p_2CrVE9%2B46nZ%40mail.gmail.com&forum_name=xsltforms-support

actions scripting using XQuery and its extensions (XUF,...) XForms recommendation is now specifying an XML notation for actions and I suggest that allowing an xf:script/@type="application/xquery" action would permit to write more complex actions with a well-known syntax. In fact, every action could be rewritten as a script instruction. Because XSLTForms already allows the author to define its own XPath functions, this would also permit to call them as actions, for browser interactivity for example.

Alain Couthures: It might make extensions easier.
Steven Pemberton: XForms uses XML Events and this gives you already an extension point. XML Events would allow it, but you would <xhtml:script type="application/xquery"> and you would be completely conformant to XML Events. XML Events is deliberately vague about what a handler is, as it covers various XML applications markup for handlers.
Nick van: [irc] http://www.w3.org/TR/xml-events2/#section-script-element
Steven Pemberton: Thank you, Nick.

Steven Pemberton: As for the second issue on HTML5 element support, I can see how HTML5 elements support what XForms already does, in a direct way.
Alain Couthures: The input types.
Steven Pemberton: Like color pickers?
Alain Couthures: Month. And items about menus.
Steven Pemberton: That sounds like it kicks off your discussion about binding controls to datatypes rather than concrete elements for the controls.
Alain Couthures: Yes.
Steven Pemberton: How should we take this forward? I'd like it if you could add a wiki page about a script extension for XQuery, showing examples of how to use it.
Kurt Cagle: What namespace?
Steven Pemberton: Not xforms namespace, but I can't see why we can't use the xhtml namespace. It should work with XML Events. Current browsers should recognize the type attribute.

Action 2010-12-1.3: Alain Couthures to add wiki page for XForms 1.2 show use cases for xquery <xhtml:script type="application/xquery">

* Next Week's XForms 1.2 topic

Alain Couthures: I'm preparing a paper for XML Prague on JSON and it's more sophisticated now.
Leigh Klotz: http://www.w3.org/MarkUp/Forms/wiki/Json

Steven Pemberton: Kurt, you can use the XBL page at http://www.w3.org/MarkUp/Forms/wiki/XBL

Steven Pemberton: Anythign else?
Leigh Klotz: How about inputmode?
Steven Pemberton: Sounds good.
Leigh Klotz: http://www.w3.org/MarkUp/Forms/wiki/InputMode

* IRC Minutes


* Meeting Ends