Hi. I believe we do same in Firefox and I think given "pros" rationale is valid, AT should see same thing as sighted users do (btw, styles are exposed to screen readers). Here's another example, if the content has bunch of whitespaces which are rendered as one character or not rendered at all then AT wants to deal with rendered text in this case. That's where the approach to expose rendered text goes from I think. Thanks. Alexander. On Tue, Apr 28, 2015 at 2:05 PM, Steve Faulkner <faulkner.steve@gmail.com> wrote: > Forwarding, as discussion is interesting and would be good to get this > defined interop wise. > -- > > Regards > > SteveF > HTML 5.1 <http://www.w3.org/html/wg/drafts/html/master/> > > ---------- Forwarded message ---------- > From: 'Alice Boxhall' via Chromium Accessibility < > chromium-accessibility@chromium.org> > Date: 28 April 2015 at 18:48 > Subject: Honouring text-transform styles in the a11y tree? > To: chromium-accessibility@chromium.org > > > Currently, Chromium exposes something like > > <div style="text-transform: uppercase">foo</div> > > as having content "FOO" in the accessibility tree. > > What do you all think of this? I see the arguments as follows: > > For: AT users perceive the content the same way as sighted users do - for > example, if checking an HTML document is rendered as expected. > Against: This type of style is intended purely for a visual effect (such > as readability) and is misleading to expose to the accessibility tree. > > To unsubscribe from this group and stop receiving emails from it, send an > email to chromium-accessibility+unsubscribe@chromium.org. > >
Received on Wednesday, 29 April 2015 12:06:26 UTC