Re: Chromebook DRM specification

On Tue, Apr 16, 2013 at 9:06 AM, Florian Bösch <pyalot@gmail.com> wrote:

> On Tue, Apr 16, 2013 at 6:02 PM, Mark Watson <watsonm@netflix.com> wrote:
>
>> I am actually saying exactly the opposite. I should have said "You can
>> certainly implement EME and a CDM ..." above.
>>
>
> I disagree with that assessment on three counts:
>
> 1) I cannot implement Widevine to deliver content to Chromebooks, so this
> is no path to satisfy the success criteria set forth by the HTML-WG charter
> 2) I cannot implement a CDM and have Chromebooks support it, so this is
> also no path to satisfy the success criteria.
> 3) I cannot implement a CDM compatible with the HTML-WG charter that would
> be deemed useful by the content industry, so that is also no path to
> satisfy the success criteria.
>
> Which means there is no path to an implementation of HTML-DRM in its
> entirety that would be both feasible and compatible with the HTML-WG
> charter.
>

Here are the success criteria from the HTML WG charter:

The HTML Working Group's work will be considered a success if there are
multiple independent complete and interoperable implementations of its
deliverable that are widely used.

   - Production of stable documents addressing the work items listed in the
   Deliverables <http://www.w3.org/2007/03/HTML-WG-charter#deliverables>
    section.
   - Test suites for each deliverable with conformance criteria
   - Availability of multiple, independent, interoperable implementations
   of each deliverable with conformance criteria;, as demonstrated by an
   implementation report (summarizing implementation status against the
   relevant test suite) for each testable class of product, including user
   agents
   - Availability of authoring tools and validation tools.
   - User community and industry adoption of the group deliverables.

Aside from the fact that it is rather early to be evaluating against these
success criteria, I fail to see anything in here that is rendered
impossible by your inability to implement your own Widevine server (1) or
ship a CDM of your choice in Google's browser (2) or create a RF DRM
solution acceptable to Hollywood (3). Everyone except Google is in the same
position as you with respect to (1) and (2) and AFAIK today, everyone is in
the same position wrt (3).

We are defining an extension point that is in some way like <object> but
with greatly restricted functionality and with the expectation that
browsers control and curate the extensions they support. These extensions
may implement DRM, but we are not proposing to standardize a complete DRM
solution.

...Mark