W3C

- DRAFT -

User Agent Accessibility Guidelines Working Group Teleconference

14 Jul 2011

See also: IRC log

Attendees

Present
Jan, Simon, Jim, Jeanne, Kelly, Kim
Regrets
Greg, Mark
Chair
jimallan, kellyFord
Scribe
kford, jeanne

Contents


<trackbot> Date: 14 July 2011

<kford> Scribe: kford

JA: Making some announcements.

Talking about web TV meeting in LA.

Survey http://www.w3.org/2002/09/wbs/36791/20110712/

JA: We tried something new by putting IER in a survey.

<JAllan> examples for 1.9.6 Martin is browsing the web with a screen reader and navigates to a web page that launches a movie with an embedded media player. The media player takes focus from the user agent and places it on a progress bar that allows the user to jump forward and back to any position in the media. Martin asks his screen reader to report where he is in the movie and is told that he is two...

<JAllan> ...minutes into a 30 minute clip.

JA: Looking at examples for 1.9.6.

<JAllan> Bob, who has a motor impairment that impacts his use of the keyboard. The web page he is use has an iframe with content that contains 75 links. Rather than moving through all of the links in the iframe using the 'tab' key, he uses his browsers 'jump to next link in the main page' key. The focus immediately jumps to the next available link after the iframe.

JA: Do these look reasonable to people?

KP: Seem reasonable to me.

<JAllan> Bob, who has a motor impairment that impacts his use of the keyboard. The web page he is using has an iframe with content that contains 75 links. Rather than moving through all of the links in the iframe using the 'tab' key, he uses his browsers 'jump to next link in the main page' key. The focus immediately jumps to the next available link after the iframe.

JR: A little confused by the Martin example.

<JAllan> kf: screenreader has a command to retrieve information from an element that has focus

JR: I thought the classic example would be when the user agent has to take focus.

<JAllan> kf: remove martin example.

Martin is off the train. Bye Martin.

JR will write an example to cover his use case.

<jeanne> ACTION: Jeanne to add the example of Bob from minutes of 14 July to 1.9.6. [recorded in http://www.w3.org/2011/07/14-ua-minutes.html#action01]

<trackbot> Created ACTION-585 - Add the example of Bob from minutes of 14 July to 1.9.6. [on Jeanne F Spellman - due 2011-07-21].

<scribe> ACTION: JR to create example for his use case around focus as discussed in 1.9.6 [recorded in http://www.w3.org/2011/07/14-ua-minutes.html#action02]

<trackbot> Created ACTION-586 - Create example for his use case around focus as discussed in 1.9.6 [on Jan Richards - due 2011-07-21].

JA: Looking at 1.9.7.

JR: L I didn't write anything because this was confusing.

s/JR:l/JR:/

JA taking an action to try and make sense of 1.9.7

<jeanne> 1.9.7 Return Focus:

<jeanne> An embedded user agent is responsible for notifying the embedding user agent that active input focus should move back to it. (Level A)

<JAllan> ACTION: jallan to make sense of 1.9.7 and report back to the group [recorded in http://www.w3.org/2011/07/14-ua-minutes.html#action03]

<trackbot> Created ACTION-587 - Make sense of 1.9.7 and report back to the group [on Jim Allan - due 2011-07-21].

Group trying to figure this out.

SH: This seems to be saying that an embedded/small user agent is responsible for telling that the larger user agent needs to take focus back.

<jeanne> An embedded (child) user agent is responsible for notifying the embedding (parent) user agent that active input focus should move back to the parent agent. (Level A)

JS: What does this accomplish, what's the accessibility angle?

SH: If the parent user agent doesn't realize that focus is returning from something that was embedded, it could return focus back to the top of the doc as an example as if this was the first interaction/focus on the page.

KP: This is a problem for example when you paste to a Google doc. You end up doing extra key presses.

<JAllan> JS: this is a point of regard issue.

<JAllan> kf: the parent needs to know it handed off the focus, and remember where it was, so when the focus return, it is placed at the previous point of regard

<JAllan> kf: to handle focus changes appropriately between parent and child user agents.

JS: I'm not sure this is testable.

Correction, JS made statement around testability.

JA: This sounds like we could go deep on this.

KP: I'll get Greg on the phone and we'll talk about this.

<scribe> ACTION: KP to resolve 1.9.7 with Greg based on discussion around parent/child user agents and ensure use cases are addressed. [recorded in http://www.w3.org/2011/07/14-ua-minutes.html#action04]

<trackbot> Created ACTION-588 - Resolve 1.9.7 with Greg based on discussion around parent/child user agents and ensure use cases are addressed. [on Kimberly Patch - due 2011-07-21].

JA: Next we talk about Patrick's comments about the DOM.
... Patrick's rewrite on 4.1.2 added structure.

<jeanne> scribe: jeanne

Kelly: does that mean that if it uses an accessibility API, then I don't have to give access to the DOM?

JA: We weren
... trying to get rid of the DOM, we were trying to broaden it to technologies that did not use a DOM.

JR: Maybe it is better to leave the term DOM, but put a note that says that you may not call it a DOM, but if you have something that works like this, then it applies.

SH: The relationship between elements is more about structure. How do I show the relationship between something disparate in the DOM, but contiguous in the rendered content?

JA: Everyone is happy with 4.1.2
... Patrick is recommending getting rid of 4.1.5, and generize 4.1.2 or 4.1.4

JR: We have lost all the write access part of the SC. It used to be.

KF: I don't wanat to lose the "if you have a DOM, give access to the AT" because that is the richest source of information for the AT.
... there is a base level for accessibility. Then we add structural, but there are other things that will enhance accessibililty that are only available from the DOM today.

SH: If you keep an internal representation, then you expose it to AT.

<JAllan> from public working draft: 2.1.5 Write Access: If the user can modify the state or value of a piece of content through the user interface (e.g., by checking a box or editing a text area), the same degree of write access is available programmatically. (Level A)

So we keep 4.1.2, 4.1.4, and 4.1.5 as is, reject Patricks comments, and change the name of 4.1.5

<kford> JA: can we come up with a name for 4.1.5?

scribe: IBM members a few years back insisted that we had to have write access. Jim will check back with Kathy (sp?).

<kford> JR: There are six items here. Are we really asking for this many or can these be condensed?

<kford> SH: I don't think 4.1.5 should be a level a.

<kford> SH and JR are going to look at this.

<kford> ACTION: Check with IBM on history here. [recorded in http://www.w3.org/2011/07/14-ua-minutes.html#action05]

<trackbot> Sorry, couldn't find user - Check

<kford> ACTION: JA check with IBM on history [recorded in http://www.w3.org/2011/07/14-ua-minutes.html#action06]

<trackbot> Created ACTION-589 - Check with IBM on history [on Jim Allan - due 2011-07-21].

<sharper> ACTION: sharper to 'Reformulate 4.1.5 / write access' [recorded in http://www.w3.org/2011/07/14-ua-minutes.html#action07]

<trackbot> Created ACTION-590 - 'Reformulate 4.1.5 / write access' [on Simon Harper - due 2011-07-21].

Resolution: Keep 4.1.2 and 4.1.4

Publishing document

<kford> JA: Does anyone have any issues with us publishing an updated working draft?

<kford> JA: No objections heard.

<kford> Resolved: UAAG agrees publishing next working draft is acceptable.

Resolution: Group votes to publish a working draft of UAAG 2.0 and Implementing UAAG 2.0

review comments - http://lists.w3.org/Archives/Public/w3c-wai-ua/2011JulSep/0010.html

<kford> JA: We have some comments on previous public draft of June 2010.

<kford> JA: There are edit changes I don't think we need to talk about.

<kford> JA: But there are other content changes I think we should talk about.

<JAllan> 4.1.2 Keystroke Precedence: The user has the option to specify that

<JAllan> keystrokes be processed in the following order: user agent user

<JAllan> interface, user agent extensions, content keystroke operations

<JAllan> administered by the user agent (e.g., access keys), and executable

<JAllan> content (e.g., key press events in scripts, etc.). (Level A)

<JAllan> -- UAs which allow Extensions usually do so in a cooperative way, i.e.

<JAllan> the extension *becomes* the UA and can do anything the UA can do; the

<JAllan> UA does not understand what the UA-Ex does nor can it prevent the

<JAllan> UA-Ex from doing something. Requiring such a separation isn't

<JAllan> reasonable. It is reasonable to request UA-Ex's not be rude, but it

<JAllan> isn't acceptable to demand this level of separation.

<JAllan> can't seem to find in current editors draft

<JAllan> 4.1.10 Override of UI Keyboard Commands: The user can override any

<JAllan> keyboard shortcut binding for the user agent user interface except for

<JAllan> conventional bindings for the operating environment (e.g., for access

<JAllan> to help). The rebinding options must include single-key and

<JAllan> key-plus-modifier keys if available in the operating environment.

<JAllan> (Level AA)

<JAllan> -- This seems like something that an accessibility agent should do on

<JAllan> its own - i.e. filter strokes and remap them before delivering to the

<JAllan> UA -- requiring this of a general UA is overkill and not particularly

<JAllan> useful. If a User needs this feature in a Web Browser, the user is

<JAllan> likely to need it in other applications as well, and having a single

<JAllan> place to do this (either at the OS level or at the AT level) is much

<JAllan> better for such a user than asking each app provide its own

<JAllan> facilities.

<JAllan> this maps to 2.1.4 Specify preferred keystrokes:(former 2.1.2): : The user can override any keyboard shortcut including recognized author supplied shortcuts (e.g. accesskeys) and user interface controls, except for conventional bindings for the operating environment (e.g., arrow keys for navigating within menus). (Level A)

<kford> JA: My response on this is that there are many who do not use AT that need this functionality.

<kford> JA: Anyone else have comments here?

<JAllan> There are other pieces of software that allow users to change keybinding natively.

<JAllan> kp: hard to find a11y agents that allow practical functional keybinding changes.

<JAllan> kp: it doesn't always work well when done natively. better to have a universal keybinding tool

<JAllan> kp: doesn't always work well depends on the application.

<JAllan> kp: can't always transfer keybindings-remapping to another user, or computer

<kford> KP:JA: Still not clear if you are asking for native app or extension?

<kford> KP: Issue is that isn't organized.

<JAllan> kf:so should this be in the OS, and be exportable, kp agrees

<kford> KP: I'm saying this is something that the user agent should do.

<kford> JS going over process for replying to comments.

<JAllan> ACTION: jallan to create table with comments and group responses. [recorded in http://www.w3.org/2011/07/14-ua-minutes.html#action08]

<trackbot> Created ACTION-591 - Create table with comments and group responses. [on Jim Allan - due 2011-07-21].

<JAllan> 4.2.3 Activate all event handlers: The user can, through keyboard

<JAllan> input alone, simultaneously activate all input device event handlers

<JAllan> explicitly associated with the content focus element. (Level A)

<JAllan> -- This doesn't make any sense. If there are 10 event handlers on a

<JAllan> given node, each of which expects state (which mouse button, which

<JAllan> keyboard key, which mouse wheel, what direction), there's no way to

<JAllan> provide a useful behavior for this. Nor is any application going to be

<JAllan> able to handle it.

<kford> Now looking at 4.2.3

<kford> This is now 2.6.3

<JAllan> 2.6.3 Activate all event handlers:: The user can, through keyboard input alone, simultaneously activate all input device event handlers explicitly associated with the content focus element. (Level A)

<kford> Group saying this likely doesn't make entire sense.

<kford> JA: 2.6 says we can list event handlers.

<kford> JA: Then activate them and finally activate all of them.

<kford> JA: Looking at UAAG 1.0 this talked about being able to activate all events of the same type.

<JAllan> JA: what do we do?

<JAllan> kf: need to review all 3 of these.

<kford> ACTION: KF to bring up areas of event handlers with PF. [recorded in http://www.w3.org/2011/07/14-ua-minutes.html#action09]

<trackbot> Created ACTION-592 - Bring up areas of event handlers with PF. [on Kelly Ford - due 2011-07-21].

<KimPatch> 1.9.7 should be taken out -- it's now 4.2.3

<kford> KP: 1.9.6 and 1.9.7 are dupes. They belong as 4.2.2 and 4.2.3 and you will see them there too.

<KimPatch> 1.9.6 is now 4.2.2

<kford> ACTION: JS please ensure duplicates of 1.9.6 and 1.9.7 are removed. [recorded in http://www.w3.org/2011/07/14-ua-minutes.html#action10]

<trackbot> Created ACTION-593 - Please ensure duplicates of 1.9.6 and 1.9.7 are removed. [on Jeanne F Spellman - due 2011-07-21].

<scribe> ACTION: Check if 1.9.6 and 1.9.7 are removed, and replaced by 4.2.2 and 4.2.3 [recorded in http://www.w3.org/2011/07/14-ua-minutes.html#action11]

<trackbot> Sorry, couldn't find user - Check

Summary of Action Items

[NEW] ACTION: Check if 1.9.6 and 1.9.7 are removed, and replaced by 4.2.2 and 4.2.3 [recorded in http://www.w3.org/2011/07/14-ua-minutes.html#action11]
[NEW] ACTION: Check with IBM on history here. [recorded in http://www.w3.org/2011/07/14-ua-minutes.html#action05]
[NEW] ACTION: JA check with IBM on history [recorded in http://www.w3.org/2011/07/14-ua-minutes.html#action06]
[NEW] ACTION: jallan to create table with comments and group responses. [recorded in http://www.w3.org/2011/07/14-ua-minutes.html#action08]
[NEW] ACTION: jallan to make sense of 1.9.7 and report back to the group [recorded in http://www.w3.org/2011/07/14-ua-minutes.html#action03]
[NEW] ACTION: Jeanne to add the example of Bob from minutes of 14 July to 1.9.6. [recorded in http://www.w3.org/2011/07/14-ua-minutes.html#action01]
[NEW] ACTION: JR to create example for his use case around focus as discussed in 1.9.6 [recorded in http://www.w3.org/2011/07/14-ua-minutes.html#action02]
[NEW] ACTION: JS please ensure duplicates of 1.9.6 and 1.9.7 are removed. [recorded in http://www.w3.org/2011/07/14-ua-minutes.html#action10]
[NEW] ACTION: KF to bring up areas of event handlers with PF. [recorded in http://www.w3.org/2011/07/14-ua-minutes.html#action09]
[NEW] ACTION: KP to resolve 1.9.7 with Greg based on discussion around parent/child user agents and ensure use cases are addressed. [recorded in http://www.w3.org/2011/07/14-ua-minutes.html#action04]
[NEW] ACTION: sharper to 'Reformulate 4.1.5 / write access' [recorded in http://www.w3.org/2011/07/14-ua-minutes.html#action07]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.136 (CVS log)
$Date: 2011/07/14 18:55:01 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.136  of Date: 2011/05/12 12:01:43  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/by7/by/
Succeeded: s/scroll bar/progress bar/
FAILED: s/JR:l/JR:/
Succeeded: s/JA: What/JS: What/
Succeeded: s/JA: I'm/JS: I'm/
Found Scribe: kford
Inferring ScribeNick: kford
Found Scribe: jeanne
Inferring ScribeNick: jeanne
Scribes: kford, jeanne
ScribeNicks: kford, jeanne
Present: Jan Simon Jim Jeanne Kelly Kim

WARNING: Replacing previous Regrets list. (Old list: GLowney, MarkH)
Use 'Regrets+ ... ' if you meant to add people without replacing the list,
such as: <dbooth> Regrets+ Greg, Mark

Regrets: Greg Mark
Found Date: 14 Jul 2011
Guessing minutes URL: http://www.w3.org/2011/07/14-ua-minutes.html
People with action items: check ja jallan jeanne jr js kf kp sharper

[End of scribe.perl diagnostic output]