User Agent Accessibility Guidelines Working Group Teleconference

06 Sep 2012

See also: IRC log


Jeanne, Greg_Lowney, Jim_Allan, Kim_Patch, [Microsoft], Jan
jimallan, kellyford


<trackbot> Date: 06 September 2012

<scribe> scribe: jallan

<scribe> agenda:

open item 1

<jeanne> http://lists.w3.org/Archives/Public/w3c-wai-ua/2012JulSep/0059.html

reviewing conformance strawman http://lists.w3.org/Archives/Public/w3c-wai-ua/2012JulSep/0059.html

js: want to add more to platform examples and platform applications

jr: why is mobile different from desktop

kp: keyboard is different

jr: not so much, complicated is browser in a browser on an OS
... in ATAG partial conformance - why something is impossible because of platform a11y issues.
... eg. we don't meet keyboard because device X has no keyboard

kf: issue not related to us, but people want mobile to be different, want guideline for mobile
... second challenge: defining what is a UA...the American Airlines app. - how much of UAAG does an app have to conform?
... apps that help with usability of other applications. are these UAs?

jr: a user experience mediated by platform and browser
... they have been broken up in the past. and a shared responsibility. but boundaries have blurred a11y is more distributed

kf: don't expect every app to provide mouseless browsing interface. should be getting that from the platform.
... should be able to say this is what I comply with, all the other parts should come from the platform or something else.

jr: concept of ATAG friendly. this is what I do...and I don't break other plugins, extensions, browser functions, external applications. If I do, then I must replace that function.

kf: new apps in windows 8 that live on the desktop.

kp: +1 to if you don't provide it, don't break it.
... developers not good with standard keyboard shortcuts., what are standards? what are best practices?
... want to make things easy. making it easy is hard. making shortcuts discoverable
... so what do I do.

jr: what we have is a highlevel of user requirements necessary to make the user experience accessible.
... how do you implement keyboard a11y?
... how do you not break somebody elses keyboard a11y?

kp: need to make it flexibile and forgiving,

gl: issues - what is UA, what is scope of responsibility, division of labor with OS and UA wrt a11y

<Jan> re: ATAG 2.0 Conforming vs. Compatible (http://lists.w3.org/Archives/Public/w3c-wai-au/2010JulSep/0089.html)

gl: things that are less than a full UA (apps)
... where to developers get guidance.

kp: in terms of developer and flexibility that go in the a11y bin and the future bin (anticipating users needs)

gl: need to add cell app as a use case for validating an SC, along with UA on desktop, media player, audio only, keyboard only
... see minutes from last week (summary) http://lists.w3.org/Archives/Public/w3c-wai-ua/2012JulSep/0058.html
... make sure the SC makes sense across these usecases.
... if the cell app breaks too many (number to be determined) then perhaps the app is not a UA
... do we have a current position on app as UA.

<jeanne> http://www.w3.org/WAI/UA/2012/ED-IMPLEMENTING-UAAG20-20120819/#intro-def-ua

js: in the implementing document http://www.w3.org/WAI/UA/2012/ED-IMPLEMENTING-UAAG20-20120827/
... see what qualifies as a UA. web application

gl: is the airline app a UA? it interprets w3 technology.

jr: problem with legislation, if law says everything that uses w3 tech, must provide full UA capabilities.

ja: share discussion with Rich S., saying they are single purpose services with a simple interface, and should be written so platform a11y tools can render the info.

gl: where is the dividing line between service and full UA

jr: the app hands off the a11y to platform or other tool. it doesn't provide a source or outline view. the best way to provide info is the itinerary view.

kp: what about apps interacting.

jr: right...app doesn't provide mouseless browsing but I don't break others mouseless browsing apps interacting with me

<Jan> JR: Example of copy...some apps support it natively, some break it...but others just play nice with the platform's copy functionality.

<Jan> JR: Boils down to maybe SC's are met: yes, no, NA and another option "is compatible with some other service or platform doing it"

kp: is functionality is native to OS then its covered, if its not, then we need some kind of test.

jr: speech SC. if someone else provides it,, then you are done. if you provide speech then you must meet the speech SC
... let developers choose what parts of UAAG they will comply with depending on things you do, and don't break platform a11y, or other extensions.

gl: so they wouldn't have to say anything about mouseless browsing, or do they have to test and say they have tested with some mouseless browsing extension.

jr: which one?

gl: in conformance claim...would have to be specific.

jr: system integrator could use all extensions to make sure it all works.
... would be nice if extension maker could claim conformance to small part of UAAG, and ensure they don't break anything else


kp: speech example say down 3 does down arrow and #3, but down 3 in another system does 3 down arrows

problems with extensions interacting, how to check non-interference.

js: concerned about partial conformance. want major browsers to conform with all of UAAG. want incentive for conformance

kp: how to draw line between partial or full conformance.

jr: full function browser.

gl: full compliance with A and x,y,z of AA.
... want a mechanism for purchasers and policy folks to make decisions. location of conformance claim needs to be available to the public.
... provide guidance to developers to do the right thing.

kp: any test for partial conformance.

gl: is there any level for minimal conformace

ja: could we have conformance for extensions? separate from full UA

kp: can we say, what do you have...this looks like a full browoser...then must comply with ____
... this is an app and you must comply with ____
... extension compliance would be useful for users to decide which extension to use.

gl: take UAAG ctegories of what is a UA, and delineate which SC apply to which type of UA.

js: new technologies seem to blur the categories.

<KimPatch> platform-based are full. Extensions or web-based have a choice to be full or partial -- does that work

kp: the blurry one is the webbased brwosers

gl: somethings can be both. e.g. Firebug, is an extension, but a huge UI, should be a full UA

kp: this makes big browsers full compliance.

jr: platform based or full. the app is platform based. has a webview, but is stand alone on the platform.

gl: what is handled by platform services to handled

kp: where does the airline app not comply?

jr: they could fail WCAG as webcontent.

kf: people writing apps don't want to read wcag because they are not writing conent.

jr: need a WAI document - a11y guidelines for webapps, smartphone apps, etc.
... should be partial conformance
... weight of conformance comes from gonvernments and policy folks

kp: NA vs not doing it...are two different things. airline app has no captions, because no video
... need distinction between we don't do this and we can't do this

gl: amount of code in major browser is huge. there may be places they can't comply because code is old an won't be for a time.

rewritten for a long time

jr: what if they have done part of an SC...

kp: isn't that partial compliance

jr: partial compliance...we don't do this, we don't break anybody else do it.

kp: 4 ways of conformance

1. comply

2. na

comply with 3rd party help

4. comply with exceptions, and list them

js: also does not comply
... how to handle, 700 correct and 1 incorrect. do we take conformance away...others are working on it. we should not jump in

kp: useful to have dialog...this complies except for thise things and here is why

gl: this fits will with last weeks discussion.
... 2 answers what passes and how
... modigy

pass/no pass/ pass with exception

list exceptions

<Greg> So for each SC the conformance claim can include:

<Greg> (a) Passes, Does Not Pass, Passes with Exceptions

<Greg> (b) Requirements (e.g. inclusion of an optional extension, platform option enabled, or implemented on user agents that implement a feature)

scribe: in addition to global requirement - not hinder OS a11y features

<Greg> That is in addition to the global configuration requirements listed for the entire conformance claim.

<Greg> (a) Passes, Does Not Pass, Passes with Exceptions listed

gl: each conformance claim covers exceptions, per platform, by extensions
... with a webbased UA, would specifiy per browser what you pass and don't pass

js: anyone to write a summary of this information

greg regrets for next week

action Kim to integrate gregs summary of last two weeks minutes into UAAG with jeanne

<trackbot> Created ACTION-757 - Integrate gregs summary of last two weeks minutes into UAAG with jeanne [on Kimberly Patch - due 2012-09-13].

conformance claims

leap motion - gesture device

fat iphone - 8 cubic foot space for hand motion, accurate and fast. hooks into touch api

flecksy keyboard for iphone

for blind users

fleksy keyboard - correction


Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.136 (CVS log)
$Date: 2012/09/06 18:35:21 $

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)

Found Scribe: jallan
Inferring ScribeNick: JAllan

WARNING: No "Topic:" lines found.

Default Present: Jeanne, Greg_Lowney, Jim_Allan, Kim_Patch, [Microsoft], Jan
Present: Jeanne Greg_Lowney Jim_Allan Kim_Patch [Microsoft] Jan
Regrets: mark
Found Date: 06 Sep 2012
Guessing minutes URL: http://www.w3.org/2012/09/06-ua-minutes.html
People with action items: 

WARNING: No "Topic: ..." lines found!  
Resulting HTML may have an empty (invalid) <ol>...</ol>.

Explanation: "Topic: ..." lines are used to indicate the start of 
new discussion topics or agenda items, such as:
<dbooth> Topic: Review of Amy's report

[End of scribe.perl diagnostic output]