Re: [External] WebXR API and low resolution captions Re: Research Questions Task Force (RQTF) agenda for 14 July 2021

Caption and text legibility have similar characteristics, probably
including resolution. The main difference is time synchronization and
presentation, which impacts characteristics such as time-to-decode per
line.

I did a quick search for text legibility and most articles state that use
cases for fonts are similar in 2D and VR, i.e., they assume legibility is
similar.

In the 1990s, HCI researchers noted that text legibility was less on
low-resolution screens (which were common in that era). But since VR
resolution doesn't yet match 2D resolution, I think that research that
found reduced legibility for low resolution displays will be true for
today's VR caption/text legibility.

On Thu, Jul 15, 2021 at 8:56 AM Joshue O'Connor <joconnor@w3.org> wrote:

> Hi all,
>
> [+ Research Questions TF list on cc]
>
> Thanks Frances for starting this, and Christopher (happy to hear more
> from George). I was actually just preparing a mail for Christopher on
> this. So for some background.
> I've reviewed a spec called 'WebXR Layers API Level 1'. [1] I think
> there is an accessibility related concern. [2]
>
> To capture that here:
>
> The spec provides support for various layer types used in a WebXR
> session (adds 'composition' layers). It covers rendering, view creation,
> Video Layer creation, events and WebXR Device API stuff.
>
> It's mostly technical implementation details - but one thing that jumped
> out at me is that WebXR Layers API allows the rendering/re-rendering of
> certain content at different
> resolutions. So I was thinking this may have accessibility implications
> to ensure that important
> accessibility related content doesn't get scaled down.
>
> So the question is should we feedback to the immersive web group? Do we
> need to flag this to them or suggest a note is needed to
> say that captions or other accessibility content should be privileged
> and therefore not rendered poorly in certain contexts, such as when
> there are bottlenecks when rendering, or limited bandwidth?
>
> What would also be useful to us in Research Questions would be evidence
> that captions actually need to be high resolution, or if there known
> issues with low resolution captions in certain contexts in immersive
> environments.
>
> Or indeed any other info you deem useful! Appreciated, and if you have
> any questions please let me know.
>
> Thanks
>
> Josh
>
> [1]  https://www.w3.org/TR/webxrlayers-1/
> [2] https://lists.w3.org/Archives/Public/public-apa/2021May/0040.html
> --
> Emerging Web Technology Specialist/Accessibility (WAI/W3C)
>
>