W3C

- DRAFT -

Web Applications Working Group Teleconference

28 Oct 2009

Agenda

See also: IRC log

Attendees

Present
[Microsoft], Shepazu, yngve, smaug, +358.407.37aaaa, khiitola
Regrets
Chair
SV_MEETING_CHAIR
Scribe
Travis

Contents


 

 

<trackbot> Date: 28 October 2009

<smaug> yngve?

<smaug> Zakim ??P1 is me

<scribe> scribe: Travis

<scribe> scribeNick: Travis

<smaug> khiitola: ping

shepazu: Who's going to TPAC?

I am.

smaug: No.

<khiitola> smaug: pong.

<smaug> khiitola: the telco is now, if you want to join

shepazu: mmm.

<khiitola> Thanks. Trying to decipher what's happening here. (sorry, I'm a newbie)

khiitola: Here to talk about touch/gesture at your convenience

shepazu: I've thought about putting touch/gesture into DOM L4 events.
... but have been advised to put it into its own spec.
... many companies have IP in this area and could be a minefeild.
... We're still not done with Dom L3 Events, so want to keep focus there.
... Folks are free to discuss on the list, but may not get much traction.
... khiitola does this fit your timeframe?

khiitola: Would love something sooner, but willing to live with the group's decision.

shepazu: Naturally, I'm don't have to be the editor...

khiitola: I vote for smaug

smaug: Hmm. very funny.

shepazu: Also very busy...
... Let's hit the agenda, and then come back to Touch stuff.

DOM3 Events in VoiceXML

shepazu: Spoke with International folks and VoicXML team recently.
... "Visual browser" preferred in that Working group ;-)
... They are using DOM L3 Events in an unorthedox way.
... May be able converge XML+V into D3E.
... Here's how they use D3E:
... No script (all declarative). For a document, the capture phase is used to determine a catch handler.
... Any catch handler (on a proximal event target) gets accumulated.
... Very similar to how a UA registered event handlers in the markup during parsing.
... Subtle differences: Only one handler will get invoked (not all possible handlers).
... Their algorithm selects the most appropriate event given current context.
... They asked whether it's reasonable to use D3E for their technique.
... Jury's still out.

smaug: MS has SALT... similar to XML+V -- and more similar to what visual browsers do.

shepazu: What if there was a SALT-like profile?

smaug: Opera has some support for XML+V (or used to?)
... Not sure how the visual browser and the XML+V interacted.
... To me, a visual browser implementor wouldn't consider integrating XHTML+Voice directly.

shepazu: Members appear to be neutral to the position, but don't want to see changes to D3E to accomodate XHTML+Voice.
... Will take this feedback to them.

Travis: (Confused about terminology)

<shepazu> http://www.w3.org/TR/xhtml+voice/

shepazu: Travis, willing to meet with these people informally at TPAC?

Travis: Yes.

<shepazu> http://www.ibm.com/developerworks/library/specification/x-v11spec/index.html

I18n telcon

shepazu: Internationalization (I18n)
... I18n have had issues with KeyBoard events in the past.
... Pleased to report that post-summary of the current KeyboardEvent plan was good.
... Mentioned that we return the key after all modifier processing has happened.
... This is what they wanted.
... Concerned about encodings in supplementary planes (upper character ranges) and how they are processed by JavaScript.

Travis: Is there a JavaScript language limitation, or is this a DOM L3 Events issue to solve?

shepazu: Not sure. Apple asked if there were keyboard that could produce characters in the supplementary ranges? Answer was yet.
... s/yet/yes/
... We might try normalizing the output.
... In JavaScript there are two escape sequences in a row. Alternate forms exist.
... Again, there are a few concerns, but these aren't blocking.
... My plan is to use JavaScript escape sequences, and provide an API to extract the Unicode code point from an escape sequence (or something like that).
... e.g., so that someone using Java could get the the same character, but may have to go through another step...
... I18n will review the spec in full, but they seem happy with it. Yeah!!!

Travis: How deep to recursive event handlers go (thinking of DOMNodeInserted handler that inserts a node).

smaug: Certain changes to the DOM within a mutation event listener are not allowed.
... not sure what other implementations do.

<scribe> ACTION: shepazu to add section on re-entrant event handlers (may be contingent on async/sync) [recorded in http://www.w3.org/2009/10/28-webapps-minutes.html#action01]

<trackbot> Created ACTION-426 - Add section on re-entrant event handlers (may be contingent on async/sync) [on Doug Schepers - due 2009-11-04].

Travis: This is also releated to sync/async.

-ms-styleContentLoaded

Travis: Wanting to introduce this experimental event to fill a void for detecting when the style sheets have all loaded.

shepazu: May want to contact annavk re: CSSOM for that.

detail attributes of resize event.

shepazu: If the property is not "in use" then it has random or default values.
... This reminds me of the "button" issue...
... PPK dislikes the W3C method (does not allow button combinations)
... What do you folks think about "buttons" with more of the IE behavior (a bit mask)

Travis: I'm hesitant to add this-- don't want to introduce the complexity and redundancy (and confusion?)

shepazu: I'd like to put this in the spec as a proposal--to get community feedback.

default actions triggered by programmatic dispatch

UNKNOWN_SPEAKER: e.g., programmatically create click event, dispatch it to an <A> element, and does it navigate the page?

shepazu: This makes me really nervous.

Travis: This is allowed in most browsers (I believe)

smaug: Yes, for back compat--we broke this once and got lots of complaints.

<smaug> ^^^^that was for keypress on input elements, IIRC

smaug: We distinguish between "trusted events" and "non-trusted", and give trusted events the ability to cause default actions.

<smaug> isTrusted

shepazu: Should we add this concept to the spec?
... This would be a new attribute for events ("trusted"={true/false}). Programmatic dispatch of events would always be trusted=false.

Ninja Dispatch

Travis: Create an event. Init the event. Call stopImmediatePropagation(). Now dispatch.

<scribe> ACTION: shepazu Add clarity about dispatch under the scenario of propagation has been stopped before dispatch. [recorded in http://www.w3.org/2009/10/28-webapps-minutes.html#action02]

<trackbot> Created ACTION-427 - Add clarity about dispatch under the scenario of propagation has been stopped before dispatch. [on Doug Schepers - due 2009-11-04].

5.1.1: "this event must not be triggered".

Travis: This does not accomodate the Window object.
... (e.g., defaultView)
... Should all events be allowed to be captured on the defaultView?

shepazu: Yes.

<scribe> ACTION: shepazu Change text in 5.1.1 to "If an event listener for the load event is attached to a node other than Window (defaultView), Document or Element nodes, or if it is attached to the bubbling phase only, this event listener must not be triggered. [recorded in http://www.w3.org/2009/10/28-webapps-minutes.html#action03]

<trackbot> Created ACTION-428 - Change text in 5.1.1 to "If an event listener for the load event is attached to a node other than Window (defaultView), Document or Element nodes, or if it is attached to the bubbling phase only, this event listener must not be triggered. [on Doug Schepers - due 2009-11-04].

Travis: That's all I have

shepazu: Want to discuss touch/gesture events?

khiitola: Would love to get a list of people interested in contributing to a specification on touch/gesture?

shepazu: It's in our charter to handle this.
... Call to advisory commitee has already gone out.
... Can ask for "interest" in the advisory committee (using official channels)
... Can ask specific people as desired.
... Note: No telecon next week.

<scribe> ACTION: shepazu to the COMM teams and PLH to make a call for participation for multitouch. [recorded in http://www.w3.org/2009/10/28-webapps-minutes.html#action04]

<trackbot> Created ACTION-429 - The COMM teams and PLH to make a call for participation for multitouch. [on Doug Schepers - due 2009-11-04].

shepazu: For multitouch, does each touchpoint have an id?

smaug: A stream id.

shepazu: What about two different people touching the device?

(Travis not participating)

shepazu: You CAN detect it!!
... Can refer to this as "alternative input events" (AIE)

khiitola: Should include the word "ninja"

<shepazu> trackbot, end telcon

Summary of Action Items

[NEW] ACTION: shepazu Add clarity about dispatch under the scenario of propagation has been stopped before dispatch. [recorded in http://www.w3.org/2009/10/28-webapps-minutes.html#action02]
[NEW] ACTION: shepazu Change text in 5.1.1 to "If an event listener for the load event is attached to a node other than Window (defaultView), Document or Element nodes, or if it is attached to the bubbling phase only, this event listener must not be triggered. [recorded in http://www.w3.org/2009/10/28-webapps-minutes.html#action03]
[NEW] ACTION: shepazu to add section on re-entrant event handlers (may be contingent on async/sync) [recorded in http://www.w3.org/2009/10/28-webapps-minutes.html#action01]
[NEW] ACTION: shepazu to the COMM teams and PLH to make a call for participation for multitouch. [recorded in http://www.w3.org/2009/10/28-webapps-minutes.html#action04]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.135 (CVS log)
$Date: 2009/10/28 20:37:30 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.135  of Date: 2009/03/02 03:52:20  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/XML+V/XHTML+Voice/
Found Scribe: Travis
Inferring ScribeNick: Travis
Found ScribeNick: Travis
Default Present: [Microsoft], Shepazu, yngve, smaug, +358.407.37aaaa, khiitola
Present: [Microsoft] Shepazu yngve smaug +358.407.37aaaa khiitola
Agenda: http://lists.w3.org/Archives/Public/www-dom/2009OctDec/0078.html

WARNING: No meeting chair found!
You should specify the meeting chair like this:
<dbooth> Chair: dbooth

Found Date: 28 Oct 2009
Guessing minutes URL: http://www.w3.org/2009/10/28-webapps-minutes.html
People with action items: shepazu

[End of scribe.perl diagnostic output]