- From: Gregg Vanderheiden <gregg@raisingthefloor.org>
- Date: Wed, 11 May 2016 16:00:10 -0500
- To: alands289 <alands289@gmail.com>
- Cc: Sailesh Panchang <sailesh.panchang@deque.com>, "Patrick H. Lauke" <redux@splintered.co.uk>, GLWAI Guidelines WG org <w3c-wai-gl@w3.org>, "public-mobile-a11y-tf@w3.org" <public-mobile-a11y-tf@w3.org>
- Message-Id: <38717FD0-C6AB-4F9D-9AB0-5476DEF386FE@raisingthefloor.org>
Correct The question is — if these are BELOW the radio button in question — is there any accessibility problem at all. ?? gregg > On May 11, 2016, at 3:20 PM, ALAN SMITH <alands289@gmail.com> wrote: > > Great information and discussion. > > Another example I see a lot of is the selection of one radio button in a form will display different additional form fields that were not previously displayed. For example: selecting insurance type radio buttons of Medical or Dental will each have their own set of sub form fields that were not displayed prior to selecting either of them. > > Alan > > Sent from Mail <https://go.microsoft.com/fwlink/?LinkId=550986> for Windows 10 > > From: Sailesh Panchang <mailto:sailesh.panchang@deque.com> > Sent: Wednesday, May 11, 2016 4:07 PM > To: Patrick H. Lauke <mailto:redux@splintered.co.uk> > Cc: w3c-wai-gl@w3.org <mailto:w3c-wai-gl@w3.org>; public-mobile-a11y-tf@w3.org <mailto:public-mobile-a11y-tf@w3.org> > Subject: Re: New SC relating to notifications of content change (was Re: Some thinking around the orientation discussion) > > Here is a proposal : > Level AA SC:"Changes in content on a Web page made by auto updates or > as a result of user action that convey information or indicate an > action are made with user awareness unless the user has opted to turn > off notification of such changes". > > This may cover change in content: > 1. based on filter / sort selections of data already displayed on page or > 2. addition(+/-) to cart or > 3. a notification that 'support by chat' is available for this task at hand, or > 4. results of form submission when they are displayed on same page > 5. a global error message placed above the form saying "form > submission failed etc." or a thank you message after completion of a > multi-step process, or > 6. on switching from grid view to list view, or > 7. in data table when sort column is changed, or > 8. on selecting a different pagination link > > Above are illustrative. > > The Intent doc of Understanding doc should clarify that it does not > cover changes like: > change in content as a result of a user selecting Tab C instead of > Tab A or opening / closing a menu as these are addressed by 4.1.2 > Nor will it cover clicking a link or button that opens up a dialog or > tooltip. These are already covered. > > Thanks, > Sailesh > > > > On 5/11/16, Sailesh Panchang <sailesh.panchang@deque.com> wrote: > > Patrick, > > About 2 step interaction: It is not me who is asking this as you > > wrongly conclude. > > this is what G80 (and H32 / H84) recommend. In suggesting this, my > > purpose is to point to one method in which 3.2.2 can be met. Notifying > > user of the expected behavior is another. > > Sailesh > > > > > > On 5/11/16, Patrick H. Lauke <redux@splintered.co.uk> wrote: > >> On 10/05/2016 16:03, Sailesh Panchang wrote: > >>> Hello Patrickk, > >>> Yes, for 3.2.2 the notification of expected behavior needs to precede > >>> the UI component. > >>> Yes, the Go-button is an older paradigm. > >>> But UI designers need to realize the accessibility challenge they > >>> create. And implementing one of these two choices will change the UI > >>> visually but help accessibility and perhaps usability too. Surely they > >>> can do something else (that almost certainly may involve a UI design > >>> change) as long as they do not pose these challenges. > >> > >> Taking the Go button case though, you're not simply asking for a visual > >> change in the UI - you're asking for an interaction change. You're > >> asking developers not to use a one-click/one-tap method that works well > >> for the majority of their users (simply activating a checkbox/radio > >> button to dynamically filter search/catalogue results) and instead > >> implementing a two-step method (activating the checkbox/radio button, > >> then pressing Go). It's a much harder sell. > >> > >>> About search results being silently displayed on the same page after > >>> activating Go button : Yes the user needs a notification say with > >>> aria-live / alert and maybe an updated heading or table caption etc. > >>> If suitable, even moving focus to that content. > >>> This is akin to error messaging when the presence of a global error > >>> message above the form is not exposed to an SR. > >> > >> And this brings us back to the point of this thread: WCAG does not have > >> a provision/SC for this sort of thing. > >> > >>> Visual proximity of updated content may not matter to SR users but it > >>> does matter generally as well as for specific PWD user groups. > >> > >> I didn't say that it didn't matter. I said that proximity cannot be used > >> as a determining factor exactly *because* it doesn't matter for all > >> users (e.g. SR users), so it would not be a suitable clause to be used > >> in SC wording. > >> > >>> I agree it is a challenge testing different device sizes, but it is > >>> just > >>> that. > >>> Usability and accessibility are in reality platform and device size > >>> specific. Something may work on laptop and responsively say, on > >>> phones / tablets of certain sizes but not on other sized phones and > >>> tablets. > >> > >> I don't dispute that it's a challenge and a reality. But again, this > >> comes down to having universally testable and determinable clauses in > >> SCs. I would argue that having an SC which may pass on one screen size > >> but fail on another - i.e. the pass/fail determination is completely > >> dependent on the auditor's actual device - is a highly subjective and > >> brittle basis for an SC that is guaranteed to make the SC completely > >> useless and uninforceable in practice. "But your honour, when I tested > >> this site on all our devices, it passed..." > >> > >>> When application / content owner is made aware of this, they need to > >>> address it if it matters to them. > >> > >> But for that to happen they need consistent and testable criteria to > >> base their assessment/fixes on. Again, having something that is > >> device-specific is not the way to go (see also the whole discussion on > >> touch target sizes in "mm as measured on screen", or large text in > >> "real-world points as measured on screen"). > >> > >> P > >> -- > >> Patrick H. Lauke > >> > >> www.splintered.co.uk | https://github.com/patrickhlauke > >> http://flickr.com/photos/redux/ | http://redux.deviantart.com > >> twitter: @patrick_h_lauke | skype: patrick_h_lauke > >> > >> > >
Received on Wednesday, 11 May 2016 21:00:39 UTC