Re: [office-accessibility] ODF TC accessibility summary from meetingearlier today

[Michael Brauer - Sun Germany - ham02 - Hamburg <Michael.Brauer@Sun.COM>]

Dave Pawson wrote On 06/15/06 12:08,:
> On 15/06/06, Michael Brauer - Sun Germany - ham02 - Hamburg
> <Michael.Brauer@sun.com> wrote:
> 
>> I see to two issues with the above proposal:
>>
>> 1. The OpenDocument specification in section 9.3 states
>>
>> "In general, an application must not render more than one of the content
>> elements contained in a frame."
>>
>> This means, an application would either render the image, or the 
>> caption, but
>> it would not render both.
> 
> 
> An example of prior art?
> 
> http://www.docbook.org/tdg/en/html/imageobject.html
> 
> <mediaobject>
>   <imageobject>
>     <imagedata fileref="figures/eiffeltower.eps" format="EPS"/>
>   </imageobject>
>   <imageobject>
>     <imagedata fileref="figures/eiffeltower.png" format="PNG"/>
>   </imageobject>
>   <textobject>
>     <phrase>The Eiffel Tower</phrase>
>   </textobject>
>   <caption>
>    <para>Designed by Gustave Eiffel in 1889, The Eiffel Tower is one of the
> most widely recognized buildings in the world.
> </para>
>   </caption>
> </mediaobject>
> 
> 
> Many media have both an image and a brief narrative for context provision?
> What is the rationale for this restriction please?

I think I was a lillte but unclear here: The <draw:frame> element specifies a 
rectangle in which content is displayed, and it is assumed that the content 
entrirely fills the rectangle. In particular, it is the frame that specifies 
a size, but not the content. So, if a <draw:frame> element contains two 
contents, then an application shall not display them both (for instance by 
displaying them both with half the height), but it shall display one or the 
other.

A caption, as it exists as a feature in office applications, is displayed in 
addition to content of the frame, but outside frame's rectangle. It therefore 
behaves differnt than the other elements inside the frame element that 
specify a content.


> 
> 
> 
> 
>>
>> 2. The caption element does not contain any information where the caption
>> will be rendered (above the frame, or below, which width will it have, 
>> etc.).
>> Actually, this would also be the case if the caption element would be
>> contained in the image element.
> 
> 
> If presentational information is to be contained within the source xml
> rather than a styling preference, yes a location attribute would
> provide this.
> I can't provide a source level example, but for a styling based 
> customisation
> see
> http://www.sagehill.net/docbookxsl/Captions.html

I think whether a caption element or the "describedBy" attribute is the 
better solution depends on the intended feature. If you use the "describedBy" 
attribute, a caption in an office application is just some regular office 
content that is defined (or marked-up) to be the caption of a certain image 
or other graphical object. No further information where and how the caption 
is displayed needs to be defined. This would on the one hand provide a lot of 
flexibility (document authors actually may declare arbitrary content to be 
the caption, regardless where it is displayed), and on the other hand would 
make the caption support a mininal change. In particular, ODF 1.0 
applications can display ODF 1.1 documents correctly, and only loose the link 
between the caption and the object.

If you add a caption element, then you have to define where and how the 
caption is displayed, and you may need additional elements and attributes to 
specify this. I don't think that this would be a bad solution in general, but 
this would be a larger change, and I fear it would break the compatibility 
with ODF 1.0 documents.

Michael