Re: Netflix HTML5 player in IE 11 on Windows 8.1

On Fri, Jun 28, 2013 at 7:37 AM, Emmanuel Revah <stsil@manurevah.com> wrote:

> On 2013/06/28 04:58, John Foliot wrote:
>
>> Emmanuel Revah wrote:
>>
>>>
>>> On 2013/06/27 18:50, John Foliot wrote:
>>> > piranna@gmail.com wrote
>>> >>
>>> >> So, if Microsoft (or whatever) doesn't port their CDM to linux, what
>>> >> can we do? What solutions we can take? Blame Microsoft? Page owner?
>>> >> W3C? Ourselves for being using linux?
>>> >>
>>> >> How can we prevent to happen this?
>>>
>>>
>>> Very good question Piranna, because of course the idea of standards is
>>> to have a standard.
>>>
>>>
>> The other points, of course, is that no standard should be forced at
>> gun-point, and not all standards will solve all problems. In fact,
>> some standards might create problems. That doesn't make them any less
>> of a standard.
>>
>
>
> A standard, in this case at least, should be implementable/accessible by
> all systems.


Well, technically, of course, EME can be implemented by all systems.
Firstly because EME is just an API not a DRM. You may consider this a
technicality, but nevertheless. Secondly, because there are no technical
issues stopping any system from shipping a closed source CDM - there are
just non-technical policy decisions on the part of some systems
implementors and thirdly because EME does not require a closed source CDM -
there are just non-technical policy decisions on the part of some content
providers that mean a non-closed-source CDM wouldn't be useful for some
content today.

It's not clear to me that these non-technical policy decisions on the part
of third parties should dictate that technical work should not proceed, if
there is value in that work, even if those policy decisions are mutually
incompatible.

Put another way, vegetarians will never be customers at a steakhouse. That
doesn't mean there's anything wrong with being a vegetarian or running a
steakhouse or that either should be expected to change.


>
>
>
> [...]
>
>> For the users of the web, this means, same old same old, if the CDM is
>>> not developed for your system then too bad for you. It's possible that
>>> Microsoft will not care to port their CDM to Linux, however, it is
>>> probable that companies willing to reach out will use multiple CDMs for
>>> their content (will require more money).
>>>
>>
>> How is "will require more money" a standards problem?
>>
>
>
> Access to a standard shouldn't require a 3rd party, EME changes that. I
> understand that it may become similar to the SSL business model as in:
> Anyone can sign their own content/certificates/etc, but for it to be
> recognised by most browsers you should require service from companies...
>  Something along that line.
>

Indeed. Anyone can implement a content protection system that works with
their own content and either ship it in their own browser or try to get
other browsers to ship it. We _could_ recommend that browsers support
pluggable downloadable CDMs to facilitate this, but it's been argued that a
confusing array of installable plugins is exactly what we _don't_ want (and
I agree).


>
> IMHO, this means access to EME is going to be almost only for businesses.
> Which brings us to:
>
>
>
>  In business, you
>> need to spend money to make money, and those companies that want to
>> maximize their profit base will spend the money and invest to do so.
>> What does cost have to do with creating a standard?
>>
>
>
> There aren't only businesses on the web, not every website is there to
> make money. I'm not sure if you are aware of that, but they still exist.
> Access to the W3C standard set should not imply monetary costs.
>
> The current set of standards doesn't imply the need for money or third
> party service providers. You can spend money on a team/CMS/etc, but you can
> also just spend time. I'm still trying to figure out how I could implement
> EME without spending any money.
>
>
>
>
>  However, this still means that many systems will be excluded from the
>>> Open Web, even if users of these are willing to accept non-free
>>> controling software on their systems.
>>>
>>
>> Please, name those systems. I keep hearing this, and keep wondering
>> which OSes, outside of Linux, are going to have an issue here? And
>> even with Linux, the issue is not technical, its philosophical.
>> Standards aren't philosophies, their standards.
>>
>
>
> I've already mentioned this somewhere but here are a few examples of
> possible configurations that may never be supported (in each case the user
> has the latest version of Firefox and Chromium and is willing to accept
> CDMs on their system):
>
> - FreeBSD / Arm
> - ReactOS / amd64
> - Linux / Sparc
>
> The list goes on. These setups exist, they are not exotic, and mostly,
> today these setups have no issues with the current set of standards.
>
> Pre-emptive counter argument: Yes Flash and Silverlight may not work for
> these, but they are not standards either.


My expectation is that EME CDMs will get wider system support than Flash
and Silverlight. You're argument is that because we can't guarantee
universal system support we should do nothing. I think this is a case of
"the perfect is the enemy of the good".

...Mark


>
>
>
>
>  If your OS and/or your
>>> architecture is not supported, then you can have a compliant browser on
>>> a modern system yet still be excluded from the "Open Web".
>>>
>>
>> However... if the OS and/or architecture refuses to implement support
>> based on moral philosophies, then the problem does not exist with the
>> technology, but with the philosophy. That's a fair position to take,
>> but that should not be *my* problem if I don't prescribe to that
>> philosophical position.
>>
>
>
> As mentioned above, I'm not talking about that case.
>
>
>
>
>  EME breaks the concept of the true Open Web and replaces it with the
>>> Web
>>> of "it's better than Flash". It also adds the notion that approved
>>> hardware can be required.
>>>
>>
>> Once again, we have a conflict of interpretation: your "True Open Web"
>> doesn't necessarily define my Open Web, and I'm OK with that. Seems
>> you are the one that has a problem with my interpretation: why must I
>> march in lockstep with you? Why must I accept your definition as the
>> one true god..err... Open Web?
>>
>
>
> I have no clue what you are trying to say, unless it's that you are
> correct and others are wrong.
>
>
>
>
>  As for the blame: you can blame Microsoft, Google and Netflix.  If it
>>> becomes a W3C standard, you can blame them too.
>>>
>>> Companies and consumers and the free market can make their own choices,
>>> true, but within the "Open Web" and with EME, it will be possible to
>>> publish 100% standard and still exclude compatible user agents
>>> (intentionally or not). There will be someone to blame for that, so I
>>> guess you could blame anyone who supports EME as a W3C standard.
>>>
>>
>> If you really feel compelled to blame anyone, blame the millions of
>> users out there that just want to be able to watch their movies on
>> their devices. They care not about the philosophical battle you are
>> waging here, and they will chose solutions that meet their needs -
>> point finale. Even ardent Linux users install "mp3 players" on their
>> system, and blissfully and happily go about their day. For them, it's
>> not religion, for them, it's entertainment.
>>
>
>
> It's not part of the W3C standard, so it's not relevant to this.
>
> Also, Linux is not a philosophy, it's a kernel for an operating system.
> Many Linux users do not have the philosophical concerns you attributed to
> them. Regardless, your use of the Linux user that installs mp3 players
> completely irrelevant. That said, irrelevance seems to be a leitmotiv for
> certain replies to this ML.
>
>
>
>
> --
> Emmanuel Revah
> http://manurevah.com
>
>
>