Re: Proposal and Justification for ARIA 1.2 (Was: text role removal)

Providing the styles that were used to generate a graphic is beyond the
scope of any of those working groups, I'm afraid (although there are some
web services that will try to identify fonts from a picture).

However, this is but one small example of why it would be useful for users
to know if the content is real text or a graphic representing text: real
text has certain functions associated with it (copy & paste, inspect
styles) that are different from graphics.

In general, I agree with Matt that whether or not to announce the role of
graphical text should be left up to the AT, and ideally should be
adjustable based on user settings.  In regular reading mode, just read the
text without announcing it.  In verbose or spell-out modes, announce it as
a single image or decorative text object.

~Amelia

On 12 August 2016 at 09:49, White, Jason J <jjwhite@ets.org> wrote:

> A slight correction: it would involve CSS, SVG and ARIA, hence would lie
> within the purview of multiple working groups/task forces.
>
> *[Jason] *
>