MATF Minutes 4 February 2016

MATF Minutes 4 February 2016 link: 
https://www.w3.org/2016/02/04-mobile-a11y-minutes.html

Text of minutes:


  Mobile Accessibility Task Force Teleconference


    04 Feb 2016

See also: IRC log <http://www.w3.org/2016/02/04-mobile-a11y-irc>


    Attendees

Present
    Kim, Alistair, Kathy, David, Detlev, Jan, Marc, Jon, Jeanne
Regrets
Chair
    Kathleen_Wahlbin
Scribe
    Kim


    Contents

  * Topics <https://www.w3.org/2016/02/04-mobile-a11y-minutes.html#agenda>
     1. failure 2.5.3
        <https://www.w3.org/2016/02/04-mobile-a11y-minutes.html#item01>
  * Summary of Action Items
    <https://www.w3.org/2016/02/04-mobile-a11y-minutes.html#ActionSummary>
  * Summary of Resolutions
    <https://www.w3.org/2016/02/04-mobile-a11y-minutes.html#ResolutionSummary>

------------------------------------------------------------------------

<Kathy> https://www.w3.org/WAI/GL/mobile-a11y-tf/wiki/Failure_of_2.5.3 
<https://www.w3.org/WAI/GL/mobile-a11y-tf/wiki/Failure_of_2.5.3>

Kathy: I won't be on the week of the 18th but I'll try to have 
understanding written by then and take a look at feedback when I get back


      failure 2.5.3

Kathy: on wiki

<Detlev> Shouldn't "due to all content and functionality not being 
available" read "due to not all content and functionality being available"?

<Jan> What about "Failure of 2.5.3 Content or functionality not 
available by touch gesture when built-in assistive technology is active"

<Jan> What about "Failure of 2.5.3 Content or functionality not 
available by touch gesture when built-in assistive technology is active"

Kathy: sections under each failure where it says a better solution would 
be to -- we don't do that in any other failure that I can see. Failure 
is usually indicating what the error condition was not a better way of 
actually doing it

Jon: there are number that do that

Kathy: below examples, above resources

David: we were nervous that people would look at the failures as ways to 
do things properly and get all mixed up

Kathy: first one is the only when I saw it in. I think it's good to have 
it in there, but I'm wondering if we should have it in their

David: let's keep that and we can link to it

<David> <p>A better solution would be to provide previous and next 
controls that

<David> allow the user to scroll the carousel by tapping (or double 
tapping when

<Kathy> Add technique for "A better solution would be to provide 
previous and next controls that allow the user to scroll the carousel by 
tapping (or double tapping when platform assistive technology such as 
screen reader) is active in addition to support the single finger swipe 
left and right functionality."

<David> platform assistive technology such as screen reader) is active 
in addition

<David> to support the single finger swipe left and right functionality.</p>

<Kathy> Add technique for "A better solution would be to use a device 
independent method such as a click handler (with platform touch support) 
that allows the user to interact with the control by platforms that 
alter touch access when built-in assistive technology is used by touch 
or through a keyboard interface."

David: add as technique instead so people don't mix up failures with 
what you're supposed to do

Kathy: and the failure should be at another step in the procedure that 
says turn on the built-in assistive technology
... step two, turn it on

David: screenreader or assistive technology?

Kathy: built-in platform assistive technology such as a screen reader

Alistair: sufficient technique M027, which is pretty much the flipside 
of this -- supporting touch with gestures. We need to blend it in with that

Kathy: could put examples in that

<agarrison> Under tests "Test that for each interactive element that 
responds to touch or gesture there is equivalent functionality provided 
and all content is available when built-in assistive technology is 
enabled that changes the touch interaction mode". Changed to "Test that 
for each interactive element that responds to touch or gesture all 
content is available when built-in assistive technology is enabled that 
changes the touch interaction mode; or the[CUT]

David: remaps touch gestures and alters the touch interface

Kathy: I understand remapping, but what does altering mean?

Jon: when the assistive technology like screen reader is running you can 
touch the screen without activating an element -- you have to double tap 
for example to send equivalent tap gestures. Wording doesn't matter

Kathy: remap better

Alistair: remap isn't exactly perfect, there must be an easier way of 
putting it but it's not for today

<David> Ex 2 <p>A better solution would be to use a device independent 
method such as a

<David> click handler (with platform touch support) that allows the user to

<David> interact with the control by platforms that alter touch access when

<David> built-in assistive technology is used by touch or through a keyboard

<David> interface.</p>

Kathy: #2 and the procedures -- doesn't specify one or more, so somebody 
could read this that they have to turn on everything all at the same 
time which may not be what we intended
... want to make sure it's valid moving forward, not just screen readers

Detlev: are there others

Jeanne: could do this is a for next loop -- repeat for each individual 
platform assistive technology

Kathy: might be instances where you want to turn on more than one, 
screenreader and screen magnifier

Detlev: rephrase for each interactive element that responds to touch or 
gesture -- is there a case for also covering generic gestures which may 
not be clearly related to any interactive element -- scroll gestures 
where you swipe across the screen without hitting interactive element in 
particular, do we need to include that?

Alistair: it was the introduction text
... for the top sentence

David: were going to have to define platform assistive technology -- do 
we want to say built-in platform assistive technology, or just platform

Jon: I thought we discussed built-in but it could apply to things like 
Android you could have third-party screen readers

David: we don't want to do that

Jon: it just complicates it

Kathy: an understanding I put system assistive technology -- we need to 
agree on a term and change it throughout. What do people prefer 
built-in, system, platform

Jeanne: I like platform it's consistent with UAAG and ATAG

David: system, platform, built-in platform, what else -- operating system
... is system a shortcut for operating system?

Jon: and other document we said platform level assistive technology 
indicating that its operating system level and not some other level like 
the browser. Browser can be operating system as well

David: people can get very technical with this language. What we really 
want to say which is too long is the assistive technology that was 
included with the operating system
... that's too long, but we can put that in a definition and have a 
short term and link
... OS assistive technology? Platform excessive technology? With the 
easiest to say, most elegant looking, most comprehensive

Alistair: edge case Samsung talkback, if we bind it too heavily to the 
platform -- Android -- we may have problems there because there may not 
be another one that is provided with a platform

David: it's the Samsung version of android

Jon: we say platform level in other places -- that's a version of that
... 4.1.4 touch proposal discussion

Kathy: consistency is a good argument to say platform

I like platform -- more clear

<agarrison> Suggestion - platform's default assistive technology

<jeanne> FRom ATAG: platform accessibility service

<jeanne> A programmatic interface that is specifically engineered to 
provide communication between applications and assistive technologies 
(e.g. MSAA, IAccessible2 and UI Automation for Windows applications, 
AXAPI for Mac OS X applications, GNOME Accessibility Toolkit API for 
GNOME applications, Java Access for Java applications). On some 
platforms, it may be conventional to enhance communication

<jeanne> further by implementing a document object.

Detlev: difficult for edge cases that come in on a skin level or vendor 
level

<jeanne> From UAAG: platform accessibility service

<jeanne> A programmatic interface that is engineered to enhance 
communication between mainstream software applications and assistive 
technologies (e.g. MSAA, UI Automation, and IAccessible2 for Windows 
applications, AXAPI for Mac OSX applications, Gnome Accessibility 
Toolkit API for GNOME applications, Java Access for Java applications). 
On some platforms it can be conventional to enhance

<jeanne> communication further by implementing a DOM.

David: interesting your using the word level -- we've been bouncing 
around platform or platform level

<jeanne> UAAG Platform def: platform

<jeanne> The software and hardware environment(s) within which the user 
agent operates. Platforms provide a consistent operational environment. 
There can be layers of software in an hardware architecture and each 
layer can be considered a platform. Native platforms include desktop 
operating system (e.g. Linux, Mac OS, Windows, etc.), mobile operating 
systems (e.g. Android, Blackberry, iOS,

<jeanne> Windows Phone, etc.), and cross-OS environments (e.g. Java). 
Web-based platforms are other user agents. User agents can employ 
server-based processing, such as web content transformations, 
text-to-speech production, etc.

<jeanne> Note 1: A user agent can include functionality hosted on 
multiple platforms (e.g. a browser running on the desktop can include 
server-based pre-processing and web-based documentation).

<jeanne> Note 2: Accessibility guidelines for developers exist for many 
platforms.

Alistair: is level adding anything?

Jon: my net be provided with a platform but interfaces at a low level 
with the platform
... I still think that talkback -- injecting JavaScript but as far as 
touch interface at the platform level not at the browser level

David: I think the edge case of Samsung I think we're okay because I 
would really say they're just modifying the platform with their own 
thing attached to it

Kathy: going to be a lot of edge cases with Android. Samsung in latest 
operating system, but that could happen all over the place

David: better off saying platform rather than platform level. If 
somebody's modifying the platform, than the assistive technology needs 
to work, and needs to work on at least one platform. I think we need to 
stay with that -- android is such a disaster right now for the 
screenreader talkback -- work on one platform is sufficient

Detlev: Samsung has a platform that their shipping and Google has a platform

Jon: platform and platform level used interchangeably

David: are people good with using platform without level?

<jeanne> +1 to platform

general agreement

Kathy: changing to platform throughout including understanding

platform assistive technology

<agarrison> Changed to "Test that for each interactive element that 
responds to touch or gesture all content is available when built-in 
assistive technology is enabled that changes the touch interaction mode; 
or there is equivalent functionality provided".

<Detlev> Shall we address general gestures here?

<Detlev> Swipe to scroll, to bring in menu from display edge, etc

Kathy: under user agent might be good to say the current technology that 
remaps gesture are...

David: also give a sense of what it is in the understanding so people 
don't have to go look somewhere else

Kathy: understanding language we had a link to gestures from the 
manufacturers -- that's good to put under resources

Jeanne: I can find them

Kathy: in this failure good to have the gestures that get remapped for 
screen readers

Jeanne: link to list

Detlev: are assistive technologies taking specific gestures, things like 
swipe, or the entire interface

Kathy: some of them don't get remaps, some of them do, you can also do 
passthrough gestures, you can also write custom gestures in some 
scenarios, so I don't think we can get more specific

<Detlev> This is a bit dated by I try to update that soon: 
http://www.incobs.de/gesten.html

<Detlev> and it's German right now

Jeanne: I'll take that definition and link throughout the document and 
techniques

David: want to make sure that it's clear that were not requiring people 
to make something work by touch and a screen reader that doesn't work by 
touch otherwise

Jon: we say that in the notes
... could be met through screen keyboard, screen controller with arrow 
keys, use the on-screen thing to meet this requirement -- there are 
other ways to meet it

<David> The technique is applicable even when

<David> access via a physical keyboard interaction is present and makes 
sure that

<David> interactions can be performed through a keyboard interface using 
touch

<David> gestures rather than relying on a physical keyboard accompanying 
the touch

<David> screen device.

Jon: the whole point of this technique is say you have a slider -- key 
up up in key down events, available through keyboard but not accessible 
through touch gestures, that's why we have this. physical keyboard 
access needs to be tested separately. Two requirements here keyboard 
interface requirement that supported with touch and keyboard interface 
requirement that supported with physical...
... keyboard.

<David> his technique applies to interactive elements on platforms where 
touch screen access is

<David> provided.

<Detlev> got thrown out of the WevEx Meeting somehow...

David: so were really talking at the element level

Jon: maybe at the feature level -- say swipe -- screen level

<agarrison> under tests 3 - suggest change "active or another touch 
method is" to "active or another method is"

David: I'll continue with this paragraph and send it off.

Kathy: will continue this discussion on the list


    Summary of Action Items


    Summary of Resolutions

[End of minutes]
------------------------------------------------------------------------
Minutes formatted by David Booth's scribe.perl 
<http://dev.w3.org/cvsweb/%7Echeckout%7E/2002/scribe/scribedoc.htm> 
version 1.144 (CVS log <http://dev.w3.org/cvsweb/2002/scribe/>)
$Date: 2016/02/04 17:06:04 $