Re: WebVR and DRM

> DRM is particularly problematic as it is built on a poor engineering
foundation and has an outsized negative effect on those who are already
marginalized by the way it blocks accessibility tools and other measures
that are otherwise legally protected under fair use.

I really do have to speak up here.

As someone who has dedicated close to 20 years advocating for, and working
towards digital inclusion and an accessible web, I really must protest how
anti-DRM advocates at the W3C keep playing the "accessibility fear" card.
Lets clear up some facts (please).

First, yes, older forms of DRM did have impacts on some users of assistive
technology, due in part to earlier blunt-force attempts at restricting data
usage. Those impediments were usually part of the delivery system however
(remember the infamous Sony root-kit affair as one example), and that is a
far cry from what we are seeing here.

However, EME, the real thing that the W3C is working on (and not DRM - stop
bending the truth), is simply the API that allows for decryption of video
streams in the browser.

But here is the really important part: once the stream is legally
decrypted, ALL OF THE ACCESSIBILITY BENEFITS OF MAKING THE BROWSER THE
VIDEO PLAYER ARE PRESENT.

Visually impaired users and mobility impaired users can still access and
interact with the (accessibility programmed) video player controls. Support
files (WebVT captions, Transcripts and audio or text-based video
descriptions) are either provided *OUTSIDE* of the encrypted video stream
(i.e. never encrypted) or, when bundled into the encrypted video wrapper,
are decrypted at the same time as the video stream - prior to any 'human'
interaction with the digital files. For all of the scary hand-waving and
allegations that somehow EME is having a negative impact on accessibility,
over the 3+ years that EME has been shipping in browsers, where is *ONE*
example of a person with a disability being prevented with viewing or
otherwise enjoying a video stream from any of the major video streaming
outfits?

Still... it's a consideration, and as the W3C is the global standards group
responsible for web technologies, with both a declared mandate of inclusion
and active working groups working towards that, we needed to be sure there
were no actual issues.

I am a member of the APA WG (Accessible Platform Architecture) and so, as
part of our cross-domain review, we have been asking and pressing the EME
WG on issues related to video accessibility going back now more than 5
years. Recently, we also published the following *RESEARCH* (not innuendo
and speculation): https://www.w3.org/2017/03/eme-accessibility.html, and
the conclusion was pretty clear.

(Regarding "Daltonization" - having done some preliminary
research personally in this matter, using a common tool that most
browser-users with color issues would use - ZoomText - I was able to
"color-shift" a video stream from Netflix in my browser, addressing the
needs of a person with atypical color perception. More research is needed,
but early indications are that this is not an issue either - unless you
seek to do physical color modification to the source file, which I was able
to conclude is not necessary. That said, additional tools are still
required today to meet this need on multiple platforms and devices, but the
encryption/decryption of the video stream does not appear to be a barrier
in addressing the use-case today.)

It's heartening to see the 'mainstream' now embrace and advocate for the
digital rights of Persons With Disabilities, but when doing so, please be
sure to base your arguments on specifics and facts, and not boogie-monster
fear-mongering. It is dishonest, insulting, and makes PWD merely pawns in a
power-struggle between legitimate business use-cases and those who
seemingly cannot discern the difference between knowledge data and
streaming entertainment.

JF

On Tue, Jul 11, 2017 at 5:41 AM, Florian Bösch <pyalot@gmail.com> wrote:

> On Mon, Jul 10, 2017 at 6:42 PM, Brandon Jones <bajones@google.com> wrote:
>
>>
>>    - Occlusion is possible through careful stenciling and alpha channel
>>    use, but not easy.
>>
>> I don't think stenciling or depth testing would be possible. In order for
> it to happen in either direction, the stencil buffer or depth buffer has to
> move from one webgl context into a "secure" webgl context. This might be
> feasible on platforms that have no concept of such a context (which would
> require secure memory) and where you could achieve "resource sharing". But
> this is a workaround that does not exist on platforms with built-in
> hardware DRM support (such as apple devices, android devices, etc.) where
> data cannot move out or into the secure context from inesecure contexts
> (that's what DRM mandates after all). Therefore stenciling/depth-testing
> support would be limited to "legacy" platforms that implement the whole DRM
> "ad-hoc" and don't make strong guarantees about accessability of data. I.e.
> it's a workaround that would not work everywhere, and as such, would be
> unsuited to the Web.
>



-- 
John Foliot
Principal Accessibility Strategist
Deque Systems Inc.
john.foliot@deque.com

Advancing the mission of digital accessibility and inclusion