W3C home > Mailing lists > Public > public-canvas-api@w3.org > January to March 2011

Re: Use Case: setCaretSelectionRect - Canvas Accessibility Working Group

From: Benjamin Hawkes-Lewis <bhawkeslewis@googlemail.com>
Date: Sun, 30 Jan 2011 13:17:43 +0000
Message-ID: <AANLkTi=2B+Acqm=8mFG9SDmy6Zkg0mLbNAAc7Sr7nb_2@mail.gmail.com>
To: Charles Pritchard <chuck@jumis.com>
Cc: public-canvas-api@w3.org, Richard Schwerdtfeger <schwer@us.ibm.com>, franko@microsoft.com, david.bolter@gmail.com, public-html-a11y@w3.org, swarren@aisquared.com, tlalor@aisquared.com, dmazzoni@google.com
On Sun, Jan 30, 2011 at 6:55 AM, Charles Pritchard <chuck@jumis.com> wrote:

> http://dl.dropbox.com/u/17337752/a11y-goog-case.png

Does this illustrate a use-case or technical requirement? If so, what is it?

Indicating the link with current focus?

> Attached is a screenshot of "Google Instant" keyboard navigation;
> my caret is on the bottom-most link.
>
> When tab is used, a focus ring is drawn on various elements.
>
> drawFocusRing is not typically used for "caret" ui controls.

What do you mean? How can we even speak about "typical" usage at this point?

> setCaretSelectionRect is useful for indicating sub-regions.

What are "sub-regions"?

> http://dl.dropbox.com/u/17337752/a11y-bing-case.png
>
> Attached is a screenshot of "Bing Images", with an image selected via mouse.
>
> Though bing image search does not seem to have workable keyboard navigation,
> it's still a reasonable use case.

What's the end-user functionality here? Allow user to choose an image
link from a list of image links as part of an image search interface?

> In this area, drawFocusRing might apply to the outermost rectangle,
> and caret selection would define a 'sub-focus' on the secondary content,
> below the selected image.

Would it? Why?

> The existing whatwg model of drawFocusRing requires an x,y parameter:
> That parameter was dropped in our new, proposal.

So are you saying your proposal should restore it?

> setCaretSelectionRect adds on width and height data, as a better means
> of "hinting" to the AT software.

Better than what?

> The word caret is used, because it corresponds to 'caret' APIs on many
> platforms, including Java and Windows. It's most often used with text
> content, with "height" and "height" parameter hinting at the font-size.

Note carets are also used next to images inline with text.

> http://dl.dropbox.com/u/17337752/a11y-combobox-case.png
>
> This screen shot recreates what I see when using an AT software package
> with caret tracking.

Which package?

> Two arrows point to the area in which the caret currently has focus.
>
> This information can be a bit difficult to pull out of an input element,
> as scrollLeft and scrollWidth have mixed support with vendors (stated in
> aria-practices document).

If UAs consistently implemented:

http://www.w3.org/TR/cssom-view/#scroll-attributes

would that solve this problem?

> 'Caret' tracking is not well defined for composite elements:
>
> http://www.w3.org/TR/2010/WD-wai-aria-practices-20100916/

Does this link illustrate an end-user use-case or technical requirement?

If so, what is it?

--
Benjamin Hawkes-Lewis
Received on Sunday, 30 January 2011 13:19:18 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Sunday, 30 January 2011 13:19:18 GMT