[html4all] role attribute keywords for embedded media

Robert J Burns rob at robburns.com
Fri Aug 22 03:57:59 PDT 2008


Hi Leif,

On Aug 22, 2008, at 6:25 AM, Leif Halvard Silli wrote:

> Robert J Burns 2008-08-21 22.41:
>
>> After recent discussions with Leif and Gregory regarding role  
>> keywords
>> for embedded media, I've added some lists of keywords to the wiki  
>> page
>> Gregory started. [...] [1]
>
>
> Excellent.
>
>> I think one of the great advantages to this role approach is it
>> would really help guide authors on inputting alt text that
>> genuinely met the requirements for creating accessible content.
>
>
> And in that regard, for what the Wiki page call 'layout or
> spacer', how about role="space" instead, to signify images which
> represent space, and therefore a space character as fallback?
>
> Btw, is role=richtext smart? And where is role=text in the Wiki?
>
> You link "richtext" to the fallback requirements, it could seem.
> But I think @role is linked to the very element. To explain why I
> think it is bad, consider XHTML 2, where e.g. an H2 element could
> contain an textimage, with the same text in the fallback.
>
> 	<h2 src=image>A little heading.</h2>
>
> What role would that H2-image play then? Richtext? I guess no. But
> the moment I emphasized one word in the heading, then it should?
>
> Richtext would also be easy to misunderstand, as many simply
> consider any format where bold, italics etc is /possible/ as rich
> text. I think role=text would be good enough.

Yeah, I can see how that might cause confusion. I was not referring to  
rich text in the markup. I never use that phrase in that way because  
it is too closely linked to the non-semantics of rich text from things  
like the rich text format (RTF). Rather I was more referring to the  
styled text of the image so that even in your example from XHTML2, the  
phrase "A little heading" might be richly styled with a font that the  
author is not able to share with users and is unable to rely on its  
presence in the users environment. Even that I consider rich text.  
However, I'll change it just to text, because its shorter and its use  
should be detailed in the recommendation anyway.

Take care,
Rob

>
>
>> [1]: <http://esw.w3.org/topic/PF/XTech/HTML5/RoleAttribute#head-35490581f69f2d7a9157e4ba18f3a20a18956cac 
>> >
>



More information about the List_HTML4all.org mailing list