Re: Netflix HTML5 player in IE 11 on Windows 8.1

On Thu, Jun 27, 2013 at 12:27 PM, Matt Ivie <matt.ivie@gmail.com> wrote:

> "No, you are missing my point entirely. Microsoft ship PlayReady as part
> of Windows Media Foundation. In principle, any browser running on Windows
> can make use of the same APIs that Internet Explorer uses to play back
> protected content. The DRM is contained in the Operating System, not
> shipped with the browser. What we call a CDM in this case is the shim
> between the EME API and the platform APIs. This does rely on MS making
> those APIs publicly available, which is why I say "in principle".
> Nevertheless, we hope MS will do this and that browsers will take advantage
> of those APIs or find some other way to support the JS APIs.
>
> You are right that a CDM that implements all the capabilities of the DRM
> itself, in software, without making use of platform APIs for that purpose,
> couldn't be Free Software as it is required to be non-user-modifiable."
>
> Your first paragraph was more on point with the idea that browsers running
> in the Windows environment could make us of the APIs on Windows systems.
>
> The second paragraph was more vague and can be interpreted as expecting
> the same behaviour in all Operating Systems.
>

I didn't intend it to be interpreted that way.


> Taken in context with only the first paragraph one could draw the
> conclusion that you're only talking about the Windows environment but taken
> in context with the broader conversation it ties into ideas previously
> discussed.
>

I'm not sure what you're getting at. Could you explain ?


>
> In regard to whether running the non-free software on the OS in order to
> give an API for EME to tie into, and whether or not it is considered a fair
> exchange or a compromise:
>
> Clearly it is a matter of opinion. I think that's why this whole thing is
> such a point of debate. The problem though, is this: While the specs for
> creating a CDM might be open, there is no way to make the CDM free
> software. No matter how you slice it, the only way to make hollywood movies
> available under this scheme is by the use of some non-free software. That I
> know of, there are no other W3C standards that can't be implemented in Free
> Software. I know that to someone that would gladly trade away their
> computing freedoms for convenience and features this seems like a very
> trivial argument to have.
>

I've given a number of examples before where, in practice, you need to use
non-free software/hardware in order to get the performance necessary for
certain applications, even though the specification could be implemented
entirely in software, with some lesser performance.
s/performance/robustness/ and you have a similar situation here. I believe
the salient difference is that for the other specifications *in principle*,
if not yet in practice, one could make a totally 'free' software/hardware
stack, whereas this is not even possible in principle for DRM given a
requirement for user-modifiability (i.e. non-robustness).

Anyway, given the absence of consensus, I'd suggest that individuals be
left free to make up their own minds whether to enter into this trade or
not, that they be as informed as possible about the consequences and that -
as far as possible - we work together to ensure that that are not asked to
trade more (in terms of security, privacy etc.) than is strictly necessary.

...Mark




>
>
> On Thu, Jun 27, 2013 at 12:01 AM, Mark Watson <watsonm@netflix.com> wrote:
>
>> On Jun 26, 2013, at 10:47 PM, Matt Ivie <matt.ivie@gmail.com> wrote:
>>
>> > On Wed, 2013-06-26 at 16:43 -0700, Mark Watson wrote:
>> >>
>> >> You are right that a CDM that implements all the capabilities of the
>> >> DRM itself, in software, without making use of platform APIs for that
>> >> purpose, couldn't be Free Software as it is required to be
>> >> non-user-modifiable.
>> >
>> > This statement also assumes that the platform that the CDM is running on
>> > has APIs that support Digital Restrictions Management.
>>
>> Actually, my preceding statement assumed that. I think you highlighted
>> the wrong part.
>>
>> > If that's the
>> > case then then something running in the OS isn't Free Software either
>> > and there is a required compromise for this solution as well. You're
>> > just moving the non-free software to a different part of the system.
>>
>> Sure. I don't think I said, or even implied, anything else (though
>> whether you consider it a compromise, or a fair exchange, is a matter
>> of personal opinion.)
>>
>> ...Mark
>>
>> >
>> >
>> > --
>> > /* Free software is a matter of liberty, not price.
>> >   Visit GNU.org * FSF.org * Trisquel.info */
>> >
>>
>
>
>
> --
> /* Free software is a matter of liberty not price.
>     Visit www.GNU.org * www.FSF.org * www.trisquel.info */
>