RE: ISSUE-30 (Longdesc) Change Proposal

Charles McCathieNevile wrote:
>
> I would like to propose that the longdesc attribute from HTML 4 be
> retained in HTML 5 as an allowed attribute on images.


I would like to signal support of this proposal as well.

While the PFWG *did* note that aria-describedby might likely supplant the 
longdesc attribute, the usage & implementation of technique is sufficiently 
different that both should be made available to content creators.  I will 
point to the following web page (http://www.cssquirrel.com/comic/?comic=41) 
as an example of a caring author 'bending' aria-describedby to actually 
deliver what longdesc does natively.

The source-code is:

<img id="imgComic" src="/images/comic/cs041.png" alt="Comic" 
aria-describedby="ariaLink" 
longdesc="http://www.cssquirrel.com/comicscripts/script41.htm"/>
<a id="ariaLink" href="http://www.cssquirrel.com/comicscripts/script41.htm" 
title="Link to a transcript of this comic">Link to a transcript of this 
comic</a>However, the author *does not* want the "Link to a transcript..." link 
visible on the page (due to legitimate design considerations), and thus uses 
CSS to remove the link from the screen - what I would argue is a 
'work-around' (aka 'hack'). Continued support for longdesc would remove the 
need for content authors 'working around' the design issue that 
aria-describedby introduces.

JF



> This implies the
> following changes to the spec:
>
> at http://dev.w3.org/html5/spec/text-level-semantics.html#the-img-
> element
> img would also become interactive content with longdesc present. The
> longdesc attribute would be listed as an attribute for the element.
>
> The attribute is described already in HTML 4 [1] and the description can
> be re-used, although it should be made clear that the URI to which
> longdesc refers can be a relative reference to some part of the same
> page
> (in order to be explicit about which content is associated with the
> image), or a different page. The example, which references an image but
> appears to provide useless alt text should not be copied from HTML 4.
>
> Other sections that may change:
> 4.8.2.1.1, 4.8.2.1.2, 4.8.2.1.3 should all mention that a longdesc *may*
> be provided to provide a detailed *description* of the image, e.g. to
> help
> a person who cannot see it to find it from a description.
>
> 4.8.2.1.5 should mention it as a way to make the association between an
> image and the relevant text explicit.
>
> 4.8.2.1.6 should mention it as the preferred way to point to a
> description
> of the image if this is desired, rather than mis-using the alt attribute
> for this purpose.
>
> 4.8.2.1.9 should mention that where an image is a key part of the
> content,
> it should have sufficient text in the alt attribute to replace the
> image,
> and using the longdesc attribute for critical information is a mistake.
> However, it can be used for additional information if desired.
>
> RATIONALE:
> This has been a controversial topic. It is clear that longdesc is
> relevant
> only to a fraction of images on the Web, and that it is only provided in
> a
> few of the cases where it is actually relevant. It is also clearly
> subject
> to bogus values to a large extent (perhaps the majority of the time).
> And
> its use is relatively limited, even by those who might be expected to
> appreciate it.
>
> However, it has been implemented multiple times successfully. The fact
> that there is bad data associated might account for low overall usage,
> but
> has relatively little impact on implementations, which can readily
> choose
> to simply ignore values which are not URIs, or even to present the value
> to the user, and relatively little impact on the user, who can still
> benefit from a *good* usage.
>
> This would require conformance checking to accept the attribute as
> valid,
> and would imply maintaining the existing requirement on Authoring
> Tools[2]
> to allow the author to use this functionality. It would maintain
> conformance of HTML-4 tools and content, rather than the current
> expected
> change leaving them non-conforming.
>
> [1] http://www.w3.org/TR/REC-html40/struct/objects.html#adef-longdesc-
> IMG
> [2] http://www.w3.org/TR/ATAG-10/ makes several relevant requiremnts
>
> cheers
>
> Chaals
>
> --
> Charles McCathieNevile  Opera Software, Standards Group
>      je parle français -- hablo español -- jeg lærer norsk
> http://my.opera.com/chaals       Try Opera: http://www.opera.com

Received on Monday, 26 October 2009 18:47:10 UTC