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

Re: February 14, 2011 - Revised CRC2D

From: Charles Pritchard <chuck@jumis.com>
Date: Mon, 14 Feb 2011 11:26:53 -0800
Message-ID: <4D5981FD.1030200@jumis.com>
To: Richard Schwerdtfeger <schwer@us.ibm.com>
CC: public-canvas-api@w3.org, dmazzoni@google.com, david.bolter@gmail.com, franko@microsoft.com, Benjamin Hawkes-Lewis <bhawkeslewis@googlemail.com>, public-html-a11y@w3.org
I've tweaked the context document.

The language in drawFocusRing about closed paths was unnecessary, and 
didn't match
the behavior and code I saw across platforms.

setCaretSelectionRect did account for rotation / skew in the transform 
I edited it to allow the UA to provide the AT with a bounding box as 
well as the revised

Following up on the rest of the agenda:

On 2/13/2011 2:07 PM, Richard Schwerdtfeger wrote:
> 4. The HTML working group response to Frank Olivier's question on 
> whether to limit canvas fallback content was to not limit the fallback 
> content:
> http://lists.w3.org/Archives/Public/public-html/2011Feb/0175.html - Frank
> http://lists.w3.org/Archives/Public/public-html/2011Feb/0176.html - Ian
> http://lists.w3.org/Archives/Public/public-html/2011Feb/0178.html - Maciej
> Rich: I recommend we not submit the canvas element (attached) changes. 
> We need a group vote.
I think we're all in consensus here.

> 3. Rich Text editing supportCaret, Selection, Grammar errors, spelling
> Annotation
> - Do the user agent manufacturers agree to do further work on rich text
> editing?
> - Where we are:
> - Rich proposal for marking caret, selected content, grammar errors,
> spelling errors using HTML 5 and ARIA with modifications
> _http://lists.w3.org/Archives/Public/public-canvas-api/2011JanMar/0001.html_
> - Positioning information of content is required - per AISquared feedback
> (See minutes from last meeting)
I believe this can be handled by keeping focus on a single element, and
defining its position via drawFocusRing and/or setCaretSelectionRect.

It's not practical to provide positioning information for all content at 
one time;
such information is re-generated programmatically in response to UI events,
such as mouse movement and keyboard input.

Image map proposals for canvas have been turned down in the past.


Received on Monday, 14 February 2011 19:27:34 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:31:52 UTC