- From: Henri Sivonen <hsivonen@iki.fi>
- Date: Mon, 27 Feb 2012 09:15:54 +0200
- To: Glenn Adams <glenn@skynav.com>
- Cc: Benjamin Hawkes-Lewis <bhawkeslewis@googlemail.com>, HTML WG LIST <public-html@w3.org>, Craig Smithpeters <Craig.Smithpeters@cox.com>
On Sun, Feb 26, 2012 at 9:58 PM, Glenn Adams <glenn@skynav.com> wrote: > Do you believe that the use of DRM/CP intentionally discriminates against > accessibility features differently that non-accessibility features? If not, > then why is the possible impact on accessibility features being singled out > in this discussion? Even if one accepted the notion that DRM discriminates equally against non-accessibility and accessibility features, discriminating against accessibility features at all deserves special attention, because traditionally under copyright law, accessibility is privileged. Typically European countries have copyright acts that put limitations on copyright (i.e. the copyright holder has less say) that enable adaptations for accessibility purposes. DRM foils this, which is a recurring theme in hearings about anti-circumvention legislation. > From the perspective of those who employ DRM/CP, there is no discrimination. > Content is content regardless of the presence or absence of accessibility > features. The only discrimination intended by DRM/CP is against unauthorized > use. ... > I would also note that, because HTML5's text track facility supports the use > of out-of-band subtitle and caption tracks, then those tracks may be > delivered and used without DRM/CP while the related video/audio resource > itself is subject to DRM/CP. Thus HTML5 itself offers a way to work around > the case where accessibility features are missing or, if present, disabled > due to lack of authorization for the media resource as a whole. Suppose the content provider isn't providing a text track and has applied DRM to the audio track. Suppose that the DRM proprietor places a contractual requirements (amplified by anti-circumvention legislation) on implementors that say the unscrambled audio samples may only be sent to audio output hardware and must not be provided to other processes. This would prevent the accessibility use of sending the audio to a speech recognition system for generating captions on the fly on the client side. (Publishers telling Amazon to disable text to speech on Kindle books and this limitation getting enforced with DRM in precedent about limitations to accessibility of this nature.) -- Henri Sivonen hsivonen@iki.fi http://hsivonen.iki.fi/
Received on Monday, 27 February 2012 07:16:23 UTC