W3C home > Mailing lists > Public > public-html@w3.org > November 2007

Re: SURVEY: Accept requirement for immediate mode graphics a la canvas element?

From: Richard Schwerdtfeger <schwer@us.ibm.com>
Date: Wed, 21 Nov 2007 15:39:00 -0600
To: James Graham <jg307@cam.ac.uk>
Cc: Anne van Kesteren <annevk@opera.com>, HTML WG <public-html@w3.org>, public-html-request@w3.org
Message-ID: <OF2BD01446.1D60DC6D-ON8625739A.007461BD-8625739A.0076E6CD@us.ibm.com>

James,

Sure. In a broader view - yes <img> would have the same problem except for
the fact that images have been captured from a a variety of means (camera,
etc.) where we have no opportunity to provide the semantics other than
possibly a longdesc or alt text.

Let's take SVG or ODF Drawings for example. With these I can apply a
properties on the markup that describes defines the object and its state.
The markup get converted to a DOM which I can access and expose through a
API.
<svg:group role="menuitem" title="launch application">

With Canvas you draw and and it goes into the bit bucket. There is noway to
capture and access the semantics of what you are doing.

With a fallback, I may have a text alternative for say a map which is a set
of directions. This is fine for someone who is blind. For someone with a
cognitive impairment I may need something else.

What we are looking at longer term is a way for users to pass preferences
along with device and user agent preferences to a source or intermediary to
select alternatives. Resource meta data would be used to describe the
resource in terms of user preferences. This is a bit beyond this example
but here are some use cases:

   A user who is deaf accesses a flash video which has closed captioning in
   english but prefers a spanish version if it is available as they speak
   spanish.
   A user who is blind acces a page with an embedded google map and would
   prefer directions in HTML because he prefers text alternatives to
   visualizations.
   I have entered a noisy room and would like text alternatives to audio
   (captioning for example) .
   I have a learning impairment which makes it difficult for me to process
   a complex visualization for a subway map. I would like a x, y, z
   alternative.

I could go on but this gives you the general idea.

To summarize:

   One alternative does not fit all even though a fallback may be an
   improvement for some.
   I am concerned that we would advocate using canvas over SVG where we
   would have an opportunity to apply semantics to the base drawing,


Rich

Rich Schwerdtfeger
Distinguished Engineer, SWG Accessibility Architect/Strategist
Chair, IBM Accessibility Architecture Review  Board
blog: http://www.ibm.com/developerworks/blogs/page/schwer


                                                                           
             James Graham                                                  
             <jg307@cam.ac.uk>                                             
                                                                        To 
             11/21/2007 01:26          Richard                             
             PM                        Schwerdtfeger/Austin/IBM@IBMUS      
                                                                        cc 
                                       Anne van Kesteren                   
                                       <annevk@opera.com>, HTML WG         
                                       <public-html@w3.org>,               
                                       public-html-request@w3.org          
                                                                   Subject 
                                       Re: SURVEY: Accept requirement for  
                                       immediate mode graphics a la canvas 
                                       element?                            
                                                                           
                                                                           
                                                                           
                                                                           
                                                                           
                                                                           




Richard Schwerdtfeger wrote:
> Hi Anne,
>
> That is a start. I am not convinced that is workable today since you
could
> place almost anything on a canvas. and your alternative may depend on the
> user. I am looking at providing a vehicle to provide this infrastructure
> now in UWA but that is a long way off.
>
> Make sense?

I'm not sure I follow. The same argument applies to <img>, right? I guess
<canvas> is richer than <img> (in that it can very dynamically) which might

exacerbate some problems although I don't have the expertise to know
exactly
what the effect will be. Perhaps you could go into a little more detail
about a)
the additional problems caused by a immediate mode graphics API, b) the
reason
that the current fallback is inadequate and c) the solution you allude to
above,
to help those of us without the background in this area understand the
issues.

--
"Eternity's a terrible thought. I mean, where's it all going to end?"
  -- Tom Stoppard, Rosencrantz and Guildenstern are Dead





graycol.gif
(image/gif attachment: graycol.gif)

pic22029.gif
(image/gif attachment: pic22029.gif)

ecblank.gif
(image/gif attachment: ecblank.gif)

Received on Wednesday, 21 November 2007 21:39:47 UTC

This archive was generated by hypermail 2.3.1 : Monday, 29 September 2014 09:38:51 UTC