RE: Action-1734, Action-1708, Action-1709: Proposal for a Figure Role

Steve Faulkner [mailto:faulkner.steve@gmail.com] wrote:
>
> It is for this reason that I suggested the mapping be changed from labelledby to
> describedby, so as to avoid almost constant repetition of the figcaption content
> being announced to users (as reported), as a description is not automatically reported
> on a group role.
> My own preferred resolution was to decouple the relationship between the accname for
> figure and figcaption and instead introduce a role of caption so that figcaption content
> could be identified as just that, a caption.


Meanwhile Richard Schwerdtfeger [mailto:schwer@us.ibm.com] wrote:
>
> I spoke with Steve Faulkner and he stated captions were more being used as descriptions
> and not basic captions.

Based on the examples Steve provided, they are Captions and not Descriptions of the complex image.  For example, in his second example *if* the image required a long description, it would be closer to something like this:

     Longdesc: a photo of the main shopping street in a the small Welsh town of Crickhowell.  It shows a number of white buildings of various commercial enterprises, with rolling green hills seen in the background. The scene also shows 2 pedestrians and 2 automobiles, one parked and another driving away from the scene.
{JF notes that this is likely overkill in this example, but it does show the difference between a caption and a long description)

> Consequently He had asked that a reference to a caption be a
> reference to a description and by default this meant an aria-describedby mapping to the
> long description. In HTML Steve was going to map <caption> through an aria-describedby
> relationship.

The problem I see here is that we may now have 2 types of Accessible Description: one a brief encapsulation of the image (with text shown to all) and a second type, (here of a true description of the scene as seen by a sighted user,) which may or may not be visually present on the host page. I am actually in favor of Steve's proposal (introduce a new role of "caption") which would be a more accurate definition of the textual content contained within <figcaption>.

I think that it is also important to note that captions / <figcaption> are not (were not?) intended to be used for long descriptions, and that good authoring guidance here is critical (a point noted in the email Steve referenced as well: http://www.w3.org/2011/06/16-html-a11y-minutes.html#item01 )

JF

Received on Monday, 30 November 2015 18:44:32 UTC