- From: Henny Swan <hswan@paciellogroup.com>
- Date: Fri, 29 May 2015 07:46:40 +0100
- To: Jeanne Spellman <jeanne@w3.org>
- Cc: MATF <public-mobile-a11y-tf@w3.org>
- Message-Id: <556761ED-30C7-4122-90B2-1892607B81A5@paciellogroup.com>
Late regrets as I was out of the office yesterday. Henny — Henny Swan User Experience and Design Lead The Paciello Group Twitter: iheni Skype: ihenix > On 28 May 2015, at 20:46, Jeanne Spellman <jeanne@w3.org> wrote: > > Minutes: http://www.w3.org/2015/05/28-mobile-a11y-minutes.html > > Text of minutes: > > [1]W3C > > [1] http://www.w3.org/ > > - DRAFT - > > Mobile Accessibility Task Force Teleconference > > 28 May 2015 > > See also: [2]IRC log > > [2] http://www.w3.org/2015/05/28-mobile-a11y-irc > > Attendees > > Present > jeanne, marcjohlic, MikeShebanek, Detlev, kim, JonAvila > > Regrets > Kathy, Jan, Michael_Pluke > > Chair > Kimberly_Patch > > Scribe > jeanne > > Contents > > * [3]Topics > 1. [4]Best Practices - Understandable > 2. [5]3.6 Provide instructions for custom touchscreen and > device manipulation gestures > * [6]Summary of Action Items > __________________________________________________________ > > <trackbot> Date: 28 May 2015 > > <scribe> scribe: jeanne > > Best Practices - Understandable > > [7]https://www.w3.org/WAI/GL/mobile-a11y-tf/wiki/Understandable > _Techniques > > [7] https://www.w3.org/WAI/GL/mobile-a11y-tf/wiki/Understandable_Techniques > > <Kim> [8]http://w3c.github.io/Mobile-A11y-TF-Note/ > > [8] http://w3c.github.io/Mobile-A11y-TF-Note/ > > <jon_avila> Response from Lisa Seeman This is an early draft of > some of what you are talking about: > [9]https://w3c.github.io/coga/issue-papers/links-buttons.html > (We already have some changes to it, but I have not uploaded > them yet.) > > [9] https://w3c.github.io/coga/issue-papers/links-buttons.html > > 3.6 Provide instructions for custom touchscreen and device > manipulation gestures > > Detlev: There are hints in iOS, that allows you to supplement > instructions - at least to labels. > ... it may be iOS only > > Marc: It is iOS specific - we have been adding info the label > of the element, so that it will work on both platforms. It adds > to the text being read. > > Jon: The hint is misunderstood. It is supposed to be the result > of the action. > ... adjustable controls will automatically give directions from > iOS Voiceover > > Mike: I think that the hint should be more than the result, it > should tell the user how to perform the action. > > <jon_avila> > [10]https://developer.apple.com/library/ios/documentation/UserE > xperience/Conceptual/iPhoneAccessibility/Making_Application_Acc > essible/Making_Application_Accessible.html#//apple_ref/doc/uid/ > TP40008785-CH102-SW6 > > [10] https://developer.apple.com/library/ios/documentation/UserExperience/Conceptual/iPhoneAccessibility/Making_Application_Accessible/Making_Application_Accessible.html#//apple_ref/doc/uid/TP40008785-CH102-SW6 > > Jon: That is contrary to Apple documentation -- perhaps we > should write a Technique to describe the best practice > > Mike: We should go back to Apple to ask them to change that > documentation. I believe it is native. > > Jon: We are talking about custom instructions. > > Jeanne: Lisa's note is interesting. I think we need to say that > user agents must give instructions for actions that it > understands. The author is responsible for custom > widgets/gestures > > Mike: THe use case I see a lot is that often there is the > ability to swipe an object from off screen to on screen. The > problem is that there is not an icon to indicate that the > material is available. > > Jon: It fits into the Help success criteria in WCAG, sometimes > there are instructions when you launch an app to show the > complex swipes. > > Kim: Some apps have overlays that show you the gestures that > can be used in the app. Then there is a button that shows how > to turn it on and off. I have seen camera apps that use complex > gestures > > Mike: It is difficult to make accessible to a screenreader user > -- to link the gesture with the instructions > > Detlev: There are some examples of using ARIA for this with > delayed text@@ > > Jon: ARIA describedby can be used. The ARIA tooltip is > difficult to get working when there is no hover, but > describedby will always be spoken even if it is offscreen. > > Kim: There are some apps that use different voices. It seems > like a different layer of visual. > > Jon: That is difficult to make work -- CSS doesn't work. The > user would have to set up the different voices > > Detlev: If there are hints, you could choose to not to render > the hints. Like there is a technique to read short link text or > longer link text. > > Jon: This is a good technique that could be used in other web > apps. > > Jeanne: This seems like a user agent issue > > Jon: JAWS has different verbosity settings. There is scripting > to turn on and off JAWS hints. There is no programmatic way to > communicate that to a webapp. > ... there was a Freedom Scientific proposal to add a help > attribute to W3C, but it never went anywhere. > > <jon_avila> > [11]https://lists.w3.org/Archives/Public/w3c-wai-gl/2003JulSep/ > 0346.html > > [11] https://lists.w3.org/Archives/Public/w3c-wai-gl/2003JulSep/0346.html > > Technique for adding hints to iOS applications > > Technique for turning hints on and off > > scribe: these should also have a optional visual layer as well, > as they are not only for audio. > > This is also a good use case. Reminder that we have a use case > wiki page, please add this to it. > > Detlev: sometimes hints should be different depending on > whether or not you are using a screenreader, which could lead > to forking. It's probably not doable. > > Kim: Why can't the user specify that they want hints for > screenreader or switch use > > Detlev: I don't think it is doable, because it requires the > author to write different types of hints. > ... It is hypothetical at the moment. > > Kim: It's good for testing. > > Jeanne: There are serious privacy issues for using sniffers. > > Jon: you could use tell what the user is doing by looking to > see whether they are using keyboard events. > > <Kim> > [12]http://www.w3.org/WAI/GL/mobile-a11y-tf/wiki/Mobile_Accessi > bility_Use_Cases > > [12] http://www.w3.org/WAI/GL/mobile-a11y-tf/wiki/Mobile_Accessibility_Use_Cases > > Jon: if we put it in terms of behaviours instead of > disabilities, that is more helpful to the user and helps with > the privacy issues. > > Jeanne: I would like to see a visual indicator that there is > information off screen. > > Kim: iOS has elegent dots and a number of how many screens. > > Mike: SOme designers will reveal a corner or sliver of > something off screen to give people an idea there is more > material available to swipe. > > Detlev: an animation affordance hint that indicates there is > more material. I am not sure we want to recommend it because it > is visual only, and not available to other modes. > > Kim: or if you aren't paying attention > > Detlev: I don't think we want to ban it > > Jeanne: it could be an example in a technique - as long as it > is available in other modalities > > Jon: another one is the labels that shake when an error is made > -- the shake doesn't give you any information about the error. > > Kim: It seems like there are two techniques here -- information > vs information available. > > Jon: the shaking could be put under existing SC, make them new > failure conditions > ... could we use 1.3.3 Sensory Characteristics for instructions > such as shake or other failures > ... we have Predictable, but we don't have affordances or > discoverable > > Detlev: a small tab on the left, if you have no way of focusing > with the screenreader, you would have to make it focusable. It > would be an elegent solution that would be available to other > modalities. > > Kim: We should indicate 1.3.3 here > > JOn: 2.1.1 keyboard -- if there is a visual indicator, then it > has to be focusable. > > Dtlev: and the swipe must be keyboard accessible, but we have > already discussed this. > > Kim: What about Form Datatypes? > > Detlev: Don't use placeholder text alone, with no label. > ... is there a current WCAG technique on Placeholder Text > > Jon: Not in current WCAG, but it is in HTML > > Detlev: Placeholder text will be read when you enter the field, > but if anything is entered, then the placeholder text will not > be read after that, unless it is marked up with ARIA label. I > don't know how to do that with native apps. > > Jon: There are ways to do it like, labelfor. Talkback announces > it after the field rather than before the field. It will remain > even after something is typed. > > Kim: there are a lot of programs that have a Back button even > in the app. If they can go back and see that placeholder text. > THere are a lot of good reasons to go back, for speech users > who accidently trip an unexpected command. Lot undo, but for > visual view. > ... that could sidestep placeholder text. > > Technique for Placeholder text and correctly labeling it if it > is used. > > Detlev: Need a Technioque for ways of marking up groups of > labels because Fieldset does not work on some mobile > environments in iOS. It makes elements hard to use. > > Jeanne: I think this is a bug to file against iOS. > > Jon: Already done. But there is also an aria technique to > handle this. > > Summary of Action Items > > [End of minutes] > > >
Received on Friday, 29 May 2015 10:12:20 UTC