User Agent Accessibility Guidelines Working Group Teleconference

07 Oct 2010

See also: IRC log


+44.797.663.aaaa, +1.512.206.aabb, jallan, +1.425.895.aacc, patrickhlauke, Greg, sharper, Jeanne, kford, Kim_Patch


<trackbot> Date: 07 October 2010

zalo, aabb is really jallan

<patrickhlauke> is my microphone not working, or is it the phone bridge?

<patrickhlauke> i can hear you folks, but you can't hear me

<patrickhlauke> hmm

<patrickhlauke> ok

<scribe> scribe: jallan

Action Item Review

discussion of action items

<jeanne> close action-247

<trackbot> ACTION-247 And MH to create techniques for 4.9 by December 3 closed

close action-268

<trackbot> ACTION-268 Craft request for input on synthesized speech inclusion in the document closed

close action-269

<trackbot> ACTION-269 Answer comment saying that we think section 3.6.1 covers the comment. closed

close action-286

<trackbot> ACTION-286 Create use case for no-script etc. for implementation doc closed

close action-289

<trackbot> ACTION-289 Write a success criteria for viewport scaling (text, images) closed

close action-291

<trackbot> ACTION-291 Write SC for user ability to open a plugin or open external player. closed

close action-319

<trackbot> ACTION-319 Copy edit draft with JS to normalize document and establish conventions closed

<jeanne> http://www.w3.org/2010/03/04-ua-minutes.html#item06

stopped review at item 329 (start with this next time)

Tooltips - device agnosticism [bug 10873] http://lists.w3.org/Archives/Public/w3c-wai-ua/2010OctDec/0003.html

gl: do we need an example about tooltips

ja: no hover on touch screens

pl: there is some discussion on how to implement hover in OS

js: should not decrease functionality because it is not suppored in tablets
... would rather touchscreen develop a hover function

gl: does make sense to have an efficient UI for commonly used functions

<jeanne> GL: things that people want to do a lot should also have a pointer-only method of accomplishing it.

gl: for headpointer, and voice input etc. need a hover function

<Greg> Should we add a recommendation to include an efficient method of carrying out frequently-used tasks using each supported input mode (e.g. keyboard, mouse, touch screen, speech). Keyboard (or keyboard equivalent)” is of course the most important because keyboard equivalent makes it device independent, but the others can be recommendations.

<Greg> Having optimized methods for pointing device and touchscreen are very useful for people using head pointers, etc.

pl: been talking for years about not using mouse over, etc, nobody listened. now with iphone, they have begun to listen.
... how to rework GL 4.1 to make more device agnostic. change to 'keyboard and keyboard like interfaces (pointers, touch screen, etc)

gl: in ISO use keybard or keybard equivelent throught the ISO document

sh: no matter the interface, the os generates keyboard scan code, and sends that to the OS for activation

<Greg> The answer is that by “keyboard” we always mean “keyboard or keyboard equivalent”; however, it is true that (a) this only makes features accessible, not necessarily convenient, and (b) UAAG20 does not currently *say* that keyboard means “keyboard (or keyboard-equivalent)”. ISO 9241-171 actually uses the phrase “keyboard (or keyboard-equivalent)” throughout. I recommend we do...

<Greg> ...the same.

sh: all keys have a scan code, when you select something the scancode is sent to the OS for activation
... even mouse clicks do this

<Greg> I don't support using the term "scan code" because that really applies only to low-level keyboard emulation, but keyboard emulation can also be done at a high level (e.g. events). Also, most readers won't understand the term "scan code" whereas most would understand "keyboard (or keyboard emulator)".

pl: problem with touchscreen, functions much like a mouse, but no hover. poor heurestics-touch and smear, or have a long touch

sh: voiceover users keyboard emulation

gl: scancode only applies to low level activity

<patrickhlauke> still not sure how touchscreen navigation (scrolling, activating with a tap, long-tapping for context) can be squared with keyboard/keyboard-like language

kf: UAAG wants to be device agnostic. yet, 4.1 is all keyboard.

<patrickhlauke> and in fact, do we need guidance on mouse interfaces?

kf: how to not waterdown 4.1, yet cover nne keyboard devices
... things that happen on webpages today were not functional with smartphones a year ago.

<patrickhlauke> if i was a radical, i'd say the title of guideline 4.1 should be "ensure full access" (with all available input modalities)

kf: smartphone UA have gotten smarter (js, etc.)
... what do we say to cover present functioning and the future.

<patrickhlauke> and then split into the various prominent modalities, with explanation (e.g. what is keyboard, with definition of kbd-like modalities)

kf: topic for face to face- touch tablet
... also do we write a response.

<jeanne> ACTION: Jeanne to draft a response to the HTML5 bug on device independence, tooltip and hover functions. [recorded in http://www.w3.org/2010/10/07-ua-minutes.html#action01]

<trackbot> Created ACTION-454 - Draft a response to the HTML5 bug on device independence, tooltip and hover functions. [on Jeanne Spellman - due 2010-10-14].

November F2F

dates are November 9 & 10

Are Smart Phone Apps user agents?

gl: don't understand question. they all sit on the OS

js: some are

kf: example of traffic detail app, it is using web servicess.

gl: some apps use internet applications, others use web protocols and are user agents

<patrickhlauke> "user agent

<patrickhlauke> A user agent is any software that retrieves, renders and facilitates end user interaction with Web content."

ja: http is web app, but udp is not

<patrickhlauke> is that the cut-off point?

js: this is related to secondlife. general concensus...if something uses uri then it needs a user agent

pl: perhaps need to define 'web content' (html, css, js, flash, etc.)
... interaction with content

kf: what about twitter apps

js: it uses webconent, it facilitates interaction

pl: not a webapp, just pulling info out of a ddatabase, in an app
... may be splitting hairs.

ja: its all about splitting hairs.

<Greg> It seems that all applications--for any platforms--fits into one of three categories: stand-alone apps that don't interact with the network or use W3C protocols or formats (i.e. standalone apps); apps that use networking but not W3C protocols or formats (i.e. Internet apps); and those that do use W3C protocols or formats (i.e. Web apps). The problem is that these distinctions are usually...

<Greg> ...transparent to the user. Two apps might be entirely identical to the user, but because one uses W3C standards "under the hood" while the other does not, only one is theoretically covered by UAAG.

kf: many tweets have urls, that can be opened with full browser

pl: twitter sending url to UA is still not a UA, if the url opens in twitter, then it is a UA and should comply

<patrickhlauke> "content (web content) includes empty content

<patrickhlauke> Information and sensory experience to be communicated to the user by means of a user agent, including code or markup that defines the content's structure, presentation, and interactions [adapted from WCAG 2.0]

<patrickhlauke> empty content (which may be alternative content) is either a null value or an empty string (e.g. one that is zero characters long). For instance, in HTML, alt="" sets the value of the alt attribute to the empty string. In some markup languages, an element may have empty content (e.g., the HR element in HTML)."

<Greg> I sort of want to avoid the term "web application", because I believe UAAG is supposed to apply to, say, SVG viewers even if they don't have any built-in network or Web awareness, *because* SVG is a W3C standard.

ja: need to define 'web content' , just because something goes over the ethernet cable does not make it webcontent (skype, im, etc)

pl: pl: murky definition

js: web content = has uri/url

gl: standalone app that plays flash. but if it transfers stuff over web, using w3 protocols

pl: looking at wcag2, define webpage, but not content

<patrickhlauke> http://www.w3.org/TR/WCAG20/#webpagedef

<patrickhlauke> wcag skirted around "what is web content" and instead defines "web page" though their intro text still talks about web content

<Greg> So, my current leaning is still towards saying UAAG applies to any software that uses W3C protocols or formats. (But I agree that leaves out software that renders proprietary formats such as Flash players, unless they transfer data over using Web protocols such as using URIs.)

pl: if we don't define what we apply to, we leave ourselve open for ?criticism?

gl: but compliance is voluntary, so may not need to define.

<Greg> I hope we don't need to provide an objective definition of what UAAG applies to because it's up to a developer to decide whether or not they want to claim compliance, and up to a purchasing organization to decide for which products or product categories they want to require compliance.

pl: we don't give authoratative definition of webcontent, here is what we think it is, if you think your device, applicatoin covers this then it is a UA

<patrickhlauke> suggest saying specifically in UAAG intro to say we don't give authoritive definition of "what is web content", and that UA authors need to decide for themselves, if they want to claim compliance, if their UA is using "web content"

gl: discuss 'raising the floor" and/or NPII to see how we can leverage, or overlap

<patrickhlauke> (gonna have to dip out, not wanting to break up your interesting - though offlist - discussion ... speak soon tho)

Summary of Action Items

[NEW] ACTION: Jeanne to draft a response to the HTML5 bug on device independence, tooltip and hover functions. [recorded in http://www.w3.org/2010/10/07-ua-minutes.html#action01]
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.135 (CVS log)
$Date: 2010/10/07 18:54:11 $

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/athorative/authoratative/
Succeeded: s/definintion/definition/
Found Scribe: jallan
Inferring ScribeNick: jallan
Default Present: +44.797.663.aaaa, +1.512.206.aabb, jallan, +1.425.895.aacc, patrickhlauke, Greg, sharper, Jeanne, kford, Kim_Patch
Present: +44.797.663.aaaa +1.512.206.aabb jallan +1.425.895.aacc patrickhlauke Greg sharper Jeanne kford Kim_Patch
Found Date: 07 Oct 2010
Guessing minutes URL: http://www.w3.org/2010/10/07-ua-minutes.html
People with action items: jeanne

[End of scribe.perl diagnostic output]