Re: Clarification

On Mon, May 13, 2013 at 1:37 PM, Brendan Aragorn <gloppius@yahoo.com> wrote:

> Mark,
> I emphatically trust my UA vendor, Mozilla.  If by gatekeeper you mean
> that they are able to block or warn of CDMs they find untrustworthy I am
> quite satisfied.
>

This is up to them. There is nothing in EME that says UAs have to provide a
plugin API enabling users to download and install arbitrary CDMs and I hope
that UAs will not do this. The EME will cause a proliferation of plugins is
one of the bigger myths that as been spread about the proposal. The
intention is to *get rid of* plugins.

So, then, the only way a CDM would be usable in the HTMLMediaElement is if
the UA has explicitly chosen to provide support for it. I would also expect
them to allow the user to enable/disable it. Possibly they may want to ask
the user for permission the first time it runs for each origin. Again, it's
up to UA implementors, but some possible ways a UA could provide support
for a given CDM might be:
- the CDM is integrated directly into the UA and ships with the UA
executable
- the CDM is implemented in the platform and the UA uses public platform
APIs to access it
- the CDM is a separately downloadable software module that the UA vendor
has integrated with and vetted. The UA will only load and execute it if it
passes some kind of code integrity check.

The last one is rather unsatisfactory, since it still looks and feels like
a plugin to the user, though it has the same properties as the first two.
It might be a path forward if there is some reason the CDM cannot ship with
the UA itself.


>   As an aside I meant that I would be willing to adopt their software as
> soon as it has proven itself to be safe.  I realize that noone can "prove"
> future action in the present.  I merely meant that those with a history of
> respecting the rights of others are more likely to continue in such fashion.
>

Fair enough. So do you agree, then, that putting DRM into the hands of the
UA vendors - who clearly have such a history - is better than proprietary
plugins or proliferation of native apps ?

...Mark


>
>   ------------------------------
>  *From:* Mark Watson <watsonm@netflix.com>
> *To:* Brendan Aragorn <gloppius@yahoo.com>
> *Cc:* public-restrictedmedia@w3.org
> *Sent:* Monday, May 13, 2013 1:01 PM
> *Subject:* Re: Clarification
>
> [This one, too, should be on restricted media]
>
> On Mon, May 13, 2013 at 11:50 AM, Brendan Aragorn <gloppius@yahoo.com>wrote:
>
> Mr Adams brought to my attention that my posting was rather unclear, so  I
> am replying to it here as well.
>
> "
> Glenn,
>
> Thank you for your message.
> On your first concern, I do not mean that they should be in anyway
> excluded, I mean that they do not warrant special rules made for their sole
> benefit.  Because something uses a great deal of bandwidth does not imply
> it has a greater cultural value than a work that does not.  Shakespeare as
> an example.
>
> On your second, I only mean that as a percentage of internet content by
> simple count of each original work they are in the minority.
>
>
> If we are assigning value to things, cultural or otherwise, then
> allocating a value of 1 to every piece of content is also a rather bad way
> to do it. At least Internet bandwidth is somewhat related to how much
> people are actually watching.
>
> I'm not saying bandwidth is a good measure - hours viewed might be better
> - I'm just saying that title count is much worse. Indeed Shakespeare would
> not do very well on title count either, at least compared to cat videos,
> say.
>
>
> As an aside, my primary concerns are that past music and past and present
> software content providers, have had no ethical qualms against installing
> harmful rootkits
> that open their customers devices to other harmful software.
>
>
> EME has an advantage here, since it places responsibility for integration
> of the CDMs with the UA vendors, who will likely demand a good
> understanding of what the CDM is doing in order to be able to offer the
> traditional security and privacy guarantees that UAs offer to users.
>
>
>   Until the "premium" content providers prove by action they will not do
> this,
>
>
> Noone can *prove* that they will not do something in the future "by
> action" in the present. You should look to the fact that most content
> providers have never done this and those that did were rightly the subject
> of some anger and have not repeated it since.
>
> Furthermore, the EME proposal addresses this risk by making the UA vendors
> the gatekeepers, unlike the current situations with plugins. [Note that if
> you don't trust your UA vendor you have bigger problems every time you
> access an Internet Banking site, for example].
>
>
> I do not trust their software and so prefer my content on my dvd player
> where it cannot actively or unitntentionally harm me.
>
>
> See above, but of course it remains your choice to obtain the content on
> DVD.
>
>
>   I also believe that open source operating systems should be supported,
> especially considering that a minor majority of the internet is running on
> them, and that most of the earths poulation cannot afford the proprietary
> ones.
>
>
> Linux support would be great. Microsoft is never going to support
> Silverlight on Linux so moving from Silverlight to EME at least offers some
> opportunity. Indeed, the first deployment of EME is on a Linux-based OS
> (ChromeOS).
>
> It seems that the things you want: better security/privacy protection and
> wider OS support are actually goals of EME, so I am not sure why you are
> opposed to it ?
>
> ...Mark
>
>
>  ------------------------------
>  *From:* Glenn Adams <glenn@skynav.com>
> *To:* Brendan Aragorn <gloppius@yahoo.com>
> *Sent:* Monday, May 13, 2013 9:47 AM
> *Subject:* Re: EME
>
>
> On Mon, May 13, 2013 at 9:43 AM, Brendan Aragorn <gloppius@yahoo.com>wrote:
>
>     I understand now more fully that the CDM requirements I suggested
> earlier would be highly impractical to implement, so I withdraw them.  I am
> not a programmer.
>
>     I do believe however that the idea that the web requires "premium
> content" to be flawed, as the admittedly vast body of fine works of
> "content providers" pale when compared to the scale of the web as a whole.
>
>
> What you are suggesting is that the Web should intentionally exclude
> serving premium content providers. Given the large numbers of individuals
> that license and use premium content, this would be antithetical to actual
> practice, wouldn't it? I've heard that Netflix content accounts for a
> significant amount of overall Internet traffic.
>
>
> It is something like a filled olympic swimming pool needing a five gallon
> bucket of water.  Counting only video Youtube alone contains many times
> more oroiginal content than might be lost if EME is not implemented.
>
>
> I don't understand what this means.
>
>
>
> Brendan Aragorn"
>
>
>
>
>
>
>