14:29:32 <RRSAgent> RRSAgent has joined #backplane
14:29:32 <RRSAgent> logging to http://www.w3.org/2008/08/12-backplane-irc
14:29:36 <Charlie> thx
14:29:43 <oedipus> Meeting: Backplane XG Teleconference
14:29:45 <Charlie> zakim, this will be RWAB
14:29:45 <Zakim> ok, Charlie; I see INC_RWAB()11:00AM scheduled to start in 31 minutes
14:29:48 <oedipus> Agenda: http://lists.w3.org/Archives/Public/public-xg-app-backplane/2008Aug/0006.html
14:29:50 <Charlie> thx
14:29:53 <oedipus> Chair: Charles_Wiecha
14:29:57 <oedipus> Scribe: Gregory_Rosmaita
14:30:00 <oedipus> ScribeNick: oedipus
14:30:01 <Charlie> you the man!
14:30:04 <oedipus> Regrets: Kevin_Kelly
14:30:14 <Charlie> you bet
14:30:57 <oedipus> "the ability to cause assistive technology (AT) to speak. By using WAI-ARIA, AxsJAX is able to manipulate the DOM such that an ARIA-aware browser plus assistive technology (AT) combination will generate and receive the necessary events which cause the AT to speak what the web developer wishes to be spoken. "
14:31:15 <Charlie> yes, was just reading the AxsJAX stuff
14:31:19 <Charlie> on google code
14:31:21 <oedipus> good
14:31:23 <Charlie> this is raman's work?
14:31:28 <oedipus> of course!
14:31:30 <Charlie> sure
14:32:13 <oedipus> the key is that we are planning on using ARIA enabled tools: dojo supports ajax, and YUI also is integrating aria support
14:32:33 <Charlie> would like to understand better what that means
14:32:41 <Charlie> devil is in the details...
14:32:49 <oedipus> if it is built-in from the start, then the level of accessibilty is much higher than if an  overlay or middleware has to intervene
14:33:02 <Charlie> we might have a really nice story in the use of xforms in dojo with the ubiquity approach
14:33:15 <Charlie> i.e. providing the accessibility DOM from the outset as an authored artifact
14:33:22 <Charlie> rather than built after the fact and exposed
14:33:28 <Charlie> not sure if any of this makes sense
14:33:29 <oedipus> basically, scripted objects don't have keyboard support or navigational flow - ARIA allows an AT to interact with an embedded widget
14:34:00 <oedipus> agree re: ubiquity approach with dojo
14:34:03 <Charlie> maybe we can look at the custom control that rahul has built in yui over the last few weeks...for calendars and color picking
14:34:11 <oedipus> yes!
14:34:16 <Charlie> then talk about how it could be made accessible using aira in dojo/xforms
14:34:22 <oedipus> i also know the head of yahoo accessibility, victor tsaran
14:34:34 <Charlie> cool
14:35:32 <oedipus> basically, ARIA keeps embedded objects and widgets from being perceptual black holes, and allows user interaction with the object/widget by bi-lateral communication between the DOM, ARIA and an AT (assistive technology)
14:36:01 <oedipus> it's something that was needed 5 years ago - we didn't know how long the development process would take
14:36:32 <oedipus> ARIA in Gmail Blog: http://www.marcozehe.de/2008/08/04/aria-in-gmail-1-alerts/
14:37:04 <oedipus> we had google reps at the last ARIA sub-committee meeting to discuss ARIA issues that they have uncovered in deployment
14:37:42 <oedipus> minutes are at: http://www.w3.org/2008/08/11-pf-minutes.html
14:38:47 <Charlie> thx...will check that out, sounds like it's the kind of feedback that is very helpful
14:39:00 <Charlie> which WG is doing ARIA?
14:41:25 <oedipus> protocols & formats: http://www.w3.org/WAI/PF
14:42:02 <oedipus> http://www.w3.org/WAI/PF/aria-primer/
14:42:18 <Charlie> thx
14:42:21 <oedipus> latest editor's drafts -- made public to assist in troubleshooting
14:42:34 <oedipus> http://www.w3.org/WAI/PF/aria - the ARIA spec itself
14:43:13 <oedipus> WAI-ARIA Roadmap (editor's draft) - http://www.w3.org/WAI/PF/aria-roadmap 
14:43:33 <oedipus> WAI-ARIA Best Practices: http://www.w3.org/WAI/PF/aria-practices
14:44:21 <oedipus> just added some resources to my proto-page - weba11y.org which is maintained by becky gibson of IBM and a dojo development team leader and point-person for accessibility (a11y)
14:45:09 <Charlie> here's the link to the calendar html source that rahul will show:
14:45:12 <Charlie> http://ubiquity-xforms.googlecode.com/svn/branches/calendar/_samples/yui-custom-controls/input-calendar.html
14:45:14 <oedipus> http://www.w3.org/2005/Incubator/app-backplane/wiki/User_talk:GJR#Developers.27_Resources
14:45:21 <oedipus> thanks -- i'll check it out now
14:45:37 <Charlie> it should come up and run directly off that URL
14:46:34 <oedipus> loading as we type..
14:46:38 <Charlie> k
14:47:51 <oedipus> problem 1: there is no explicit "Submit" mechanism - this is a WCAG 1.0/2.0 violation
14:48:05 <oedipus> http://www.w3.org/TR/wcag10
14:48:12 <oedipus> http://www.w3.org/TR/wcag20
14:48:13 <Charlie> doubt this page actually does anything other than show the calendars...
14:48:18 <oedipus> ah
14:48:29 <Charlie> it's to illustrate the code as being checked into the ubiquity project
14:48:35 <Steven> Steven has joined #backplane
14:48:41 <Charlie> hi steven!
14:48:48 <oedipus> still, would be good to make sure the example is WCAG compliant
14:48:53 <oedipus> aloha, steven!
14:48:56 <Steven> hi
14:49:02 <Steven> scribing on another call right now
14:49:08 <Charlie> let's map out a demo for this XG to do that
14:49:17 <Charlie> that we can then show at the tech plenary
14:49:25 <oedipus> wunderbar!
14:49:40 <Charlie> LOL
14:50:59 <oedipus> recorded demos of ARIA in action: http://www.weba11y.com/demos/index.html
14:53:04 <Steven> üÜ for Gregory's copy buffer
14:53:18 <oedipus> thanks - er, danke
14:58:12 <Zakim> INC_RWAB()11:00AM has now started
14:58:19 <Zakim> +Gregory_Rosmaita
14:58:32 <oedipus> rrsagent, make minutes
14:58:32 <RRSAgent> I have made the request to generate http://www.w3.org/2008/08/12-backplane-minutes.html oedipus
14:59:20 <oedipus> rrsagent, set logs world-visible
14:59:27 <oedipus> rrsagent, make minutes
14:59:27 <RRSAgent> I have made the request to generate http://www.w3.org/2008/08/12-backplane-minutes.html oedipus
14:59:36 <Zakim> +[IBM]
14:59:46 <Charlie> zakim, [IBM] is Charlie
14:59:46 <Zakim> +Charlie; got it
15:00:21 <rahul> rahul has joined #backplane
15:00:53 <oedipus> present+ Rahul
15:01:07 <oedipus> present+ Peter_Thomas
15:01:53 <jackjansen> jackjansen has joined #backplane
15:03:14 <Steven> zakim, dial steven-617
15:03:17 <Zakim> ok, Steven; the call is being made
15:03:19 <Zakim> +Steven
15:04:10 <oedipus> TOPIC: Introductions & Agenda Additions/General Comments
15:04:11 <Zakim> +John_Boyer
15:04:37 <oedipus> http://lists.w3.org/Archives/Public/public-xg-app-backplane/2008Aug/0006.html
15:04:38 <Charlie> http://ubiquity-xforms.googlecode.com/svn/branches/calendar/_samples/yui-custom-controls/input-calendar.html
15:04:48 <Zakim> + +31.20.616.aaaa
15:04:54 <John_Boyer> John_Boyer has joined #backplane
15:05:02 <Charlie> http://ubiquity-xforms.googlecode.com/svn/branches/calendar/_samples/yui-custom-controls/input-calendar.html
15:05:14 <oedipus> zakim, aaaa is Jack_Jansen
15:05:14 <Zakim> +Jack_Jansen; got it
15:05:20 <oedipus> rrsagent, make minutes
15:05:20 <RRSAgent> I have made the request to generate http://www.w3.org/2008/08/12-backplane-minutes.html oedipus
15:05:33 <Charlie> Agenda: http://lists.w3.org/Archives/Public/public-xg-app-backplane/2008Aug/0006.html
15:06:16 <oedipus> CW: GJR and i have been chatting about ARIA and accessibility
15:06:33 <oedipus> CW: want to move towards planning set of demos for tech plenary
15:06:34 <Steven> rssagent, make log public
15:06:42 <Steven> rrsagent, make minutes
15:06:42 <RRSAgent> I have made the request to generate http://www.w3.org/2008/08/12-backplane-minutes.html Steven
15:06:53 <John_Boyer> rrsagent, make log public
15:06:57 <oedipus> CW: need to get back to Debi Dahl and identify other WGs we want to meet with
15:06:58 <John_Boyer> rrsagent, make minutes
15:06:58 <RRSAgent> I have made the request to generate http://www.w3.org/2008/08/12-backplane-minutes.html John_Boyer
15:07:13 <oedipus> CW: start with rahul's demo
15:07:34 <oedipus> CW: segue from custom control to ARIA stuff - custom control accessibility
15:07:54 <oedipus> CW: abstract UI control tree might support ARIA enabling
15:08:19 <oedipus> CW: JJ read paper last night - didn't know about Mark Birbeck's SMIL work
15:08:27 <oedipus> CW: scripting limitations of SMIL
15:08:34 <oedipus> CW: agenda addenda?
15:08:45 <Steven> zakim, mute me
15:08:45 <Zakim> Steven should now be muted
15:09:03 <oedipus> TOPIC: Rahul's Calendar Picker Example
15:09:12 <oedipus> http://ubiquity-xforms.googlecode.com/svn/branches/calendar/_samples/yui-custom-controls/input-calendar.html
15:09:20 <oedipus> CW: platforms? using mac
15:09:32 <oedipus> GJR: using FF3.01 with JAWS9
15:09:53 <oedipus> s/with JAWS/with JAWS9 on WinXP Pro
15:10:22 <oedipus> GJR will try later with orca (GNOME screen reader) and FF3.01
15:10:36 <John_Boyer> John_Boyer has joined #backplane
15:10:37 <oedipus> CW: shall i use web dialogs?
15:10:49 <oedipus> JJ: doesn't work in safari -- doesn't crash it, but nothing happens
15:10:59 <Charlie> http://ubiquity-xforms.googlecode.com/svn/branches/calendar/_samples/yui-custom-controls/input-calendar.html
15:11:01 <oedipus> CW: don't think will work with Safari and Opera
15:11:45 <Steven> rrsagent, make minutes
15:11:45 <RRSAgent> I have made the request to generate http://www.w3.org/2008/08/12-backplane-minutes.html Steven
15:12:29 <oedipus> Rahul: calendar control exposed as xforms input or xforms range - airline reservation scenario 
15:12:54 <oedipus> Rahul: started with YUI library - object, leverage YUI widgets and add ubiquity-xform
15:13:35 <oedipus> Rahul: 3 custom controls - reason for 3 variants is to explore xforms markup - one uses CSS class, one uses datatype and one uses ???
15:14:02 <oedipus> Rahul: could come from different AJAX libraries; using xforms model to tie together across AJAX libraries (backplane end of it)
15:14:22 <oedipus> Rahul: can do variants by tweaking appearance -- popup calendar, standing calendar
15:14:36 <oedipus> Rahul: just started with this excercise 
15:14:59 <oedipus> ??: current value in data model that you are bound to, can edit from either direction
15:15:14 <Charlie> s/??/Charlie
15:15:56 <oedipus> Rahul: since using data bindings provided by xforms model - if look at source, very little application code to perform synchronization; using AJAX library YUI provides bindings; still need application mode 
15:16:17 <oedipus> CW: using ubiquity model all UI provided by YUI framework
15:16:36 <oedipus> CW: xforms controls in play, AJAX doing custom widget rendering; child of xforms control
15:17:12 <oedipus> CW: if look at DOM, separate label for xforms, child container element for what is needed by AJAX library which perfoms rendering
15:17:40 <Steven> zakim, who is muted?
15:17:40 <Zakim> I see Steven, Jack_Jansen muted
15:17:54 <oedipus> CW: tying to accessibility, dives down below visible markup; factors DOM in running browser; interaction and data binding used for presentational purposes, will be easy to integrate ARIA interfaces
15:18:08 <oedipus> TOPIC: Approach to AJAX
15:18:31 <oedipus> CW: abstract UI control good starting point for adding ARIA to support assistive technology
15:18:58 <oedipus> CW: illustrates use case - ubiquity approach via AJAX in sub-trees, could be leveraged for accessibility
15:19:12 <oedipus> CW: any questions/comments on controls?
15:19:13 <rahul> s/???/XForms appearance attribute/
15:19:27 <oedipus> GJR: lack of explicit submit mechanism for form (WCAG violation)
15:19:38 <oedipus> JB: hangs my machine
15:20:06 <oedipus> Rahul: can try locally -- incredibly faster
15:20:30 <oedipus> JJ: if inspect dom via FF, first control has input stuff, but where is calendar input
15:20:53 <oedipus> JJ: if walk through DOM tree see input control, but no trace of the calenar inputs
15:21:26 <oedipus> Rahul: in FireBug go to HTML tag
15:21:38 <oedipus> Rahul: open BODY and will find TABLE in there; expand that
15:21:51 <oedipus> JJ: table incomplete -- only first column before variable tag
15:22:03 <oedipus> JJ: ah, hidden in different place then i thought they'd be
15:22:25 <oedipus> JJ: table with 1 row that happens to stack
15:23:47 <oedipus> CW: ARIA makes distinction between A11y DOM and UA DOM; additional level of tree stucture; why isn't this abstract UI tree in XForms?  giving authoring story to provide community 
15:26:22 <oedipus> http://www.w3.org/WAI/PF/aria/#tree
15:26:25 <oedipus> http://www.w3.org/WAI/PF/aria/#treegrid
15:27:32 <oedipus> i think i need to call back in
15:27:36 <Steven> we hear you
15:27:36 <Zakim> -Gregory_Rosmaita
15:27:58 <Charlie> k
15:28:20 <Steven> ack me
15:28:42 <Steven> zakim, mute me
15:28:42 <Zakim> Steven should now be muted
15:28:51 <Zakim> +Gregory_Rosmaita
15:29:00 <Steven> JohnBoyer: I still use IE 3.2
15:29:12 <Steven> :-p
15:29:52 <John_Boyer> I want to make sure we still have Opera compatibility!
15:32:05 <oedipus> http://www.w3.org/WAI/PF/aria/#grid
15:36:57 <oedipus> CW: think about xf namespace elements are where properties are 
15:37:21 <thomas> thomas has joined #backplane
15:37:24 <oedipus> CW: intuition - hiding calendar is clean -- does accessibilty make cleaner
15:37:48 <oedipus> CW: should we take further to ascertain how accessibility cna be obtained
15:38:02 <oedipus> JJ: like what i hear - doesn't mess up DOM, IDs all there -- much better
15:39:05 <oedipus> CW: shadowing - XBL something like this
15:39:58 <oedipus> CW: demo ideal for the point that can work today - AJAX implementation plus ARIA could provide support for xforms on today's browsers - good to take to WAI and Forms at Tech Plenary
15:40:42 <oedipus> CW: make sense to follow up with GJR - working on simple load form to illustrate simple app - not only nice looking form that runs in YUI and/or AJAX
15:41:08 <jackjansen> q+ to ask whether there's a timeframe for Safari support
15:41:10 <oedipus> ??: calendaring important - may need more work 
15:41:19 <oedipus> CW: part of GUI aspect of form
15:41:33 <oedipus> GJR: if have access to code, can ARIA enable
15:41:51 <oedipus> best ARIA test suite so far: http://test.cita.uiuc.edu/aria/
15:43:49 <oedipus> CW: get discussion going on list - can point to form fragments, and add ARIA support to them
15:44:22 <oedipus> ??: loan form - how do you tell someone who cannot see something that is not relevant and now visible?
15:45:01 <Charlie> s/??/John_Boyer
15:45:04 <oedipus> ??: if type principal for loan, not enough info - need to set duration of loan - still need to calculate interest rate; when entered enough info, can show you something you couldn't do before
15:45:16 <Charlie> s/??/John_Boyer
15:47:04 <oedipus> CW: map between XForms and ARIA
15:47:37 <oedipus> GJR: will provide list of ARIA derived from XForms
15:47:57 <oedipus> CW: mapping between ubiquity events and control - core javascript behavior in ubiquity that is ARIA enabled
15:48:38 <oedipus> TOPIC: XForms Model and UI controls in SMIL
15:49:02 <oedipus> CW: interesting notion of data model extension serving as UI for SMIL agent to drive dynamic properties and synchronize
15:49:18 <oedipus> JJ's background paper: http://lists.w3.org/Archives/Public/public-xg-app-backplane/2008Aug/0005.html
15:49:40 <oedipus> CW: wanted to highlight need for custom loop code to drive the google map API
15:50:33 <oedipus> CW: MarkB has example using XBL - can reuse that - uses data model and UI into googlemap logic; set of xforms elements that handles all google map info; on SMIL side, have xforms control as well; can use ubiquity to add databinding into bicycle tour
15:51:30 <oedipus> JJ: back up one step - idea started as way of getting state in model; evolving into SMIL execution model - declarative, unlike javascript; trying to see how a declarative execution model enable a certain classs of applicatinos without using javascript
15:52:20 <oedipus> JJ: always will be apps that need javascript, but large number of places where could use SMIL plus data model; applications can then be transformed, accessified - can do what you can't do in the javascript black hole
15:52:42 <oedipus> CW: showed in paper SMIL and XForms components
15:53:21 <oedipus> JJ: map feature - googlemaps part of problem - modifying text in HTML - if xforms output can be used, once data model changes, can change visible text on page but cannot change the URI of a link
15:53:38 <oedipus> JJ: hoping to use XBL - but no XBL for Safari
15:54:09 <oedipus> JJ: screen shots on page 2 - can't change link via xforms - so little bit of javascript that smashes value from data model into HREF of anchor
15:54:58 <oedipus> JJ: for my types of apps, if have attribute value templates - in any attribute value, can put xforms expression in curly braces and gets evaluated at runtime
15:55:17 <oedipus> CW: could have same result as rahul
15:55:36 <oedipus> JJ: then have to tell stylesheets that though this is not an anchor tag, should use formatting defined for anchor
15:55:52 <oedipus> CW: Ubiquity SMIL project Mark Birbeck is running?
15:55:54 <oedipus> JJ: no
15:56:02 <oedipus> CW: just found out about it recently
15:56:11 <oedipus> GJR: notes that safari being ARIA enabled
15:56:26 <oedipus> JJ: have to look at how much of SMIL is implemented for type of demos want to sho
15:57:23 <oedipus> JJ: would solve plugin problem when viewing demo
15:57:40 <oedipus> GJR: http://juicystudio.com/article/safari-support-aria.php
15:58:14 <oedipus> CW: will post SMIL ubiquity info on list
15:59:10 <Charlie> http://code.google.com/p/ubiquity-smil/
15:59:18 <oedipus> JJ: research work is "dumb" from one point of view; from another deployable, anything deployable should advance the work
15:59:32 <jackjansen> s/dumb/done
15:59:59 <oedipus> CW: compelling point in paper is data driven API - show how Mark done from maps side and JJ from SMIL side - add to list of tech plenary demos?
16:00:13 <oedipus> JJ: wouldn't mind doing demo, but participation in person a big if
16:01:05 <oedipus> JJ: if ubiquity-smil code can help me with more depth - video synchronization; smil plugin for safari
16:01:12 <oedipus> CW: limitations
16:01:23 <oedipus> JB: ubiquity is targetting safari as well
16:01:33 <oedipus> JB: won't be long, but might not be soon enough
16:02:10 <Zakim> -Steven
16:02:17 <oedipus> JJ: understand: FF would be nicer platform, but in Safari, get binding integration free - not available in FF - anyone know of javascript bridge for FF?
16:02:42 <jackjansen> Javascript <-> Python, that is...
16:02:56 <oedipus> CW: next week: set of demos that can take to tech plenary - calendar picker and JJ's demo
16:03:09 <Zakim> -John_Boyer
16:03:11 <Zakim> -Charlie
16:03:13 <Zakim> -Jack_Jansen
16:03:13 <John_Boyer> John_Boyer has left #backplane
16:03:15 <Charlie> Charlie has left #backplane
16:03:15 <oedipus> CW: take pulse of projects and decide about moving forward
16:03:30 <oedipus> rrsagent, make minutes
16:03:30 <RRSAgent> I have made the request to generate http://www.w3.org/2008/08/12-backplane-minutes.html oedipus
16:03:34 <Zakim> -Gregory_Rosmaita
16:03:35 <Zakim> INC_RWAB()11:00AM has ended
16:03:36 <Zakim> Attendees were Gregory_Rosmaita, Charlie, Steven, John_Boyer, +31.20.616.aaaa, Jack_Jansen
16:04:22 <oedipus> present- +31.20.616.aaaa
16:04:26 <oedipus> rrsagent, make minutes
16:04:26 <RRSAgent> I have made the request to generate http://www.w3.org/2008/08/12-backplane-minutes.html oedipus
16:06:01 <oedipus> ACTION: Gregory - post list of aria roles, properties and states derived from or related to XForms concepts
16:06:06 <oedipus> rrsagent, make minutes
16:06:06 <RRSAgent> I have made the request to generate http://www.w3.org/2008/08/12-backplane-minutes.html oedipus
16:09:04 <rahul> rahul has left #backplane
16:20:09 <oedipus> rrsagent, please part
16:20:09 <RRSAgent> I see 1 open action item saved in http://www.w3.org/2008/08/12-backplane-actions.rdf :
16:20:09 <RRSAgent> ACTION: Gregory - post list of aria roles, properties and states derived from or related to XForms concepts [1]
16:20:09 <RRSAgent>   recorded in http://www.w3.org/2008/08/12-backplane-irc#T16-06-01