RE: ACTION-1442: Draft spec text for aria-current and aria-currentfor

Aria-active ?


From: Matthew King [mailto:mattking@us.ibm.com]
Sent: Wednesday, November 19, 2014 5:43 PM
To: LWatson@PacielloGroup.com
Cc: 'Joseph Scheuhammer'; 'W3C WAI Protocols & Formats'; 'Alexander Surkov'
Subject: RE: ACTION-1442: Draft spec text for aria-current and aria-currentfor

Léonie wrote:
> The problem is that people don’t think in terms of the “displayed page” or “displayed step”,
> but we do think in terms of the “current page” and “current step”.
> That applies equally to developers and users I think.

The discussions of the definition and purpose of the attribute do not lead me to that conclusion.

I am now convinced that "displayed" is definitely worse than "current" for authors (not sure about screen reader users -- that would take a study).

Even though aria-displayed is definitely out , I still feel that aria-current is undesirably vague and could likely be the source of a great deal of confusion. The more clear the name of the attribute, the less we have to rely on a broad audience reading the specification to grasp the basic meaning.

I know we like names short, but low frequency of use and low numbers of instances per page, typically no more than one, should mitigate any concerns about length if a longer name would improve understanding.

Matt King
IBM Senior Technical Staff Member
I/T Chief Accessibility Strategist
IBM BT/CIO - Global Workforce and Web Process Enablement
Phone: (503) 578-2329, Tie line: 731-7398
mattking@us.ibm.com<mailto:mattking@us.ibm.com>



From:        Léonie Watson <LWatson@PacielloGroup.com<mailto:LWatson@PacielloGroup.com>>
To:        Matthew King/Fishkill/IBM@IBMUS, "'Alexander Surkov'" <surkov.alexander@gmail.com<mailto:surkov.alexander@gmail.com>>,
Cc:        "'Joseph Scheuhammer'" <clown@alum.mit.edu<mailto:clown@alum.mit.edu>>, "'W3C WAI Protocols & Formats'" <public-pfwg@w3.org<mailto:public-pfwg@w3.org>>
Date:        11/19/2014 12:43 PM
Subject:        RE: ACTION-1442: Draft spec text for aria-current and aria-currentfor
________________________________



“Matt's response:
Use case: You have a list of 5 nav links ("page 1", "page 2", ... "page 5") in a nav element. The  link for "page 1" has been activated so that page is displayed in the main content. The anchor for "page 1" has aria-displayed="true". When reading the content of the nav element, the screen reader user would hear something like "Link page 1 displayed".”

The problem is that people don’t think in terms of the “displayed page” or “displayed step”, but we do think in terms of the “current page” and “current step”. That applies equally to developers and users I think.

Léonie.




--
Senior Accessibility Engineer, TPG
@LeonieWatson @PacielloGroup

From: Matthew King [mailto:mattking@us.ibm.com]
Sent: 19 November 2014 19:52
To: Alexander Surkov
Cc: Joseph Scheuhammer; W3C WAI Protocols & Formats
Subject: Re: ACTION-1442: Draft spec text for aria-current and aria-currentfor

Alex wrote:
> displayed sounds like visible with me, something on the screen

Matthew King <mattking@us.ibm.com<mailto:mattking@us.ibm.com>> wrote:
> Yes, that is the understanding we want to convey, right?

Alex wrote:
> Not sure. What about navigation links when all of them are presented and only one of them is "current"?

Matt's response:
Use case: You have a list of 5 nav links ("page 1", "page 2", ... "page 5") in a nav element. The  link for "page 1" has been activated so that page is displayed in the main content. The anchor for "page 1" has aria-displayed="true". When reading the content of the nav element, the screen reader user would hear something like "Link page 1 displayed".

Matt King
IBM Senior Technical Staff Member
I/T Chief Accessibility Strategist
IBM BT/CIO - Global Workforce and Web Process Enablement
Phone: (503) 578-2329, Tie line: 731-7398
mattking@us.ibm.com<mailto:mattking@us.ibm.com>



From:        Alexander Surkov <surkov.alexander@gmail.com<mailto:surkov.alexander@gmail.com>>
To:        Matthew King/Fishkill/IBM@IBMUS,
Cc:        Joseph Scheuhammer <clown@alum.mit.edu<mailto:clown@alum.mit.edu>>, "W3C WAI Protocols & Formats" <public-pfwg@w3.org<mailto:public-pfwg@w3.org>>
Date:        11/19/2014 11:28 AM
Subject:        Re: ACTION-1442: Draft spec text for aria-current and aria-currentfor
________________________________




Not sure. What about navigation links when all of them are presented and only one of them is "current"?

On Wed, Nov 19, 2014 at 2:15 PM, Matthew King <mattking@us.ibm.com<mailto:mattking@us.ibm.com>> wrote:
Alex wrote:
> displayed sounds like visible with me, something on the screen

Yes, that is the understanding we want to convey, right?

Matt King
IBM Senior Technical Staff Member
I/T Chief Accessibility Strategist
IBM BT/CIO - Global Workforce and Web Process Enablement
Phone: (503) 578-2329<tel:%28503%29%20578-2329>, Tie line: 731-7398
mattking@us.ibm.com<mailto:mattking@us.ibm.com>



From:        Alexander Surkov <surkov.alexander@gmail.com<mailto:surkov.alexander@gmail.com>>
To:        Matthew King/Fishkill/IBM@IBMUS,
Cc:        Joseph Scheuhammer <clown@alum.mit.edu<mailto:clown@alum.mit.edu>>, "W3C WAI Protocols & Formats" <public-pfwg@w3.org<mailto:public-pfwg@w3.org>>
Date:        11/19/2014 11:12 AM
Subject:        Re: ACTION-1442: Draft spec text for aria-current and aria-currentfor
________________________________




displayed sounds like visible with me, something on the screen

On Wed, Nov 19, 2014 at 1:57 PM, Matthew King <mattking@us.ibm.com<mailto:mattking@us.ibm.com>> wrote:
After further work and thought, I am still questioning the name.
Couldn't we create greater clarity and understanding by calling this aria-displayed?

1. When it comes to screen reader users, if I heard "displayed", I think I would more easily figure out what it means.

2. When it comes to authors, I would think it would be much easier to understand the difference between displayed and selected vs current and selected.

3. I also would not be surprised if "displayed" resulted in more accurate translations by vendors of non-english screen readers.

I understand there is apossibility it would more narrowly scope its application, but that could be a really good thing when it comes to creating clarity. However, I have not yet heard of a use case where the meaning was not synonomous with "displayed".

Matt King
IBM Senior Technical Staff Member
I/T Chief Accessibility Strategist
IBM BT/CIO - Global Workforce and Web Process Enablement
Phone: (503) 578-2329<tel:%28503%29%20578-2329>, Tie line: 731-7398
mattking@us.ibm.com<mailto:mattking@us.ibm.com>



From:        Matthew King/Fishkill/IBM@IBMUS
To:        Joseph Scheuhammer <clown@alum.mit.edu<mailto:clown@alum.mit.edu>>,
Cc:        public-pfwg@w3.org<mailto:public-pfwg@w3.org>
Date:        11/19/2014 10:22 AM
Subject:        Re: ACTION-1442: Draft spec text for aria-current and aria-currentfor
________________________________




Revisions to yesterday's proposal:

1. Clarified that false and undefined are equivalent and false is the default.

2. Clarified that UA should not expose and AT should not convey false or undefined.

In the following text, phrases in square brackets are intended to be links.

Aria-current attribute

Indicates an element represents the current item within a container or set of related elements.

The aria-current attribute indicates whether an element represents what is current (true), or not current (false). If the aria-current attribute is false or undefined, User Agents SHOULD NOT expose the aria-current state of an element, and assistive technologies SHOULD NOT convey it.

The aria-current attribute is used when one of the elements in a set of related elements has a visual style different from other members in the set to indicate that the element identifies what is current. For example, it can be used to indicate which [link] in a set of [navigation] links is visually styled to indicate that it is the link for the currently displayed page. Similarly, it can be used to indicate which step in a [list] of wizard steps is visually styled to inform the user that the currently displayed wizard content is for that step.

Note:
When applied to an element contained in a widget that supports selection, the meaning of aria-current is different from the meaning of [aria-selected]. Authors should not use aria-current in lieu of aria-selected and should avoid using aria-current in circumstances where the meaning of aria-current would be the same as aria-selected. For example, in a single-select [tablist] where the selected [tab] element corresponds to the displayed tabpanel, aria-current is unnecessary. However, if the selected state of a tab is used to indicate which tab is selected for an action, such as move, delete, or display (make current), then aria-current should be used to indicate which tab represents the currently displayed tabpanel. Examples that further explain how to use aria-current and aria-selected are available in the [WAI-ARIA Authoring Practices].

Characteristics of aria-current

Used in roles: All elements of the base markup.

Value: true/false

Values of aria-current

true: The element is current.
false (default): The element is not current

Matt King
IBM Senior Technical Staff Member
I/T Chief Accessibility Strategist
IBM BT/CIO - Global Workforce and Web Process Enablement
Phone: (503) 578-2329<tel:%28503%29%20578-2329>, Tie line: 731-7398
mattking@us.ibm.com<mailto:mattking@us.ibm.com>



From:        Joseph Scheuhammer <clown@alum.mit.edu<mailto:clown@alum.mit.edu>>
To:        Matthew King/Fishkill/IBM@IBMUS, Joseph Scheuhammer <clown@alum.mit.edu<mailto:clown@alum.mit.edu>>,
Cc:        public-pfwg@w3.org<mailto:public-pfwg@w3.org>
Date:        11/19/2014 09:58 AM
Subject:        Re: ACTION-1442: Draft spec text for aria-current and aria-currentfor
________________________________




Léonie, Matt,

Thanks for your clarifications.

> So, what is the best way to write the text so the undefine/false
> equivalency is clear?

For the "Values" table, how about:

"false (default):  The element is not current.
true: The element is current."

I'm following the style of the value tables for other boolean aria-*
attributes. For example, see the table for aria-disabled
(http://rawgit.com/w3c/aria/master/aria/aria.html#aria-disabled).

Similarly, for the "Value" entry in the "Characteristics" table:

"Value: true/false"

Also, given that "conveyed by User Agents" has to do with the
accessibility API, I suggest this re-wording:

"If the aria-current attribute is false or undefined, User Agents SHOULD
NOT expose the aria-current state of an element, and assistive
technologies SHOULD NOT convey it."

Hope that helps.

--
;;;;joseph.

'Array(16).join("wat" - 1) + " Batman!"'
         - G. Bernhardt -