- From: Leif Halvard Silli <xn--mlform-iua@xn--mlform-iua.no>
- Date: Mon, 21 Mar 2011 08:45:33 +0100
- To: Steve Faulkner <faulkner.steve@gmail.com>
- Cc: HTML Accessibility Task Force <public-html-a11y@w3.org>, HTMLWG WG <public-html@w3.org>
Steve Faulkner, Sun, 20 Mar 2011 13:05:00 -0700:
> I have written up example text adding it to the current spec text for
> image. The additional text is identified using the <ins> element and
> is futher differentiated from the current spec text using a
> background color.
>
> http://www.html5accessibility.com/tests/img-longdesc.html
Question: Why did you conclude that @longdesc should make img
interactive content?
]] If the element has a usemap or longdesc attribute:
Interactive content. [[
*That* was an error in the previous change proposal. Why? Because HTML5
forbids nesting of interactive content. [1] Which means that two of the
use cases that @longdesc is claimed useful for in HTML4, would be
forbidden:
<!--would be forbidden:-->
<a href=* ><img longdesc=* src=* alt=* ></a>
<!--would be forbidden as it would represent 2 interactivisms-->
<img longdesc=* src=* alt=* usemap=#* >
I've said this before (but perhaps not in public-html): HTML5 has
really cut the corners when it comes to nesting of interactive content.
In HTML4 and XHTML1, there e.g. was no restriction on wrapping a link
around an image map: <a href=*><img src=* alt=* usemap=#*></a>. But
this is not permitted in HTML5 anymore.
@longdesc should be considered more like @cite of <blockquote> - which
aren't interactive content. Or what if there will be an
@aria-describedAT="URL" at some point in the future? Should *that* also
make the element interactive?
I hope you can update your experimental spec text to reflect this.
[1]
http://dev.w3.org/html5/spec/introduction.html#restrictions-on-content-models-and-on-attribute-values
--
leif halvard silli
Received on Monday, 21 March 2011 07:47:15 UTC