EME as originally conceived is about security for vendors, yes. But Firefox and Chrome, which take their roles as user agent seriously, made sure the spec is one that they can also implement securely on their end. This is a ridiculously huge improvement over NPAPI.
Browser plugins can restrict your freedom in two ways:
1. They can monitor and restrict what you do with the content you're accessing through the plugin.
2. They can monitor and restrict what you do with ANYTHING ELSE IN THE BROWSER OR ANYWHERE ON YOUR COMPUTER.
EME-restricted video can do 1, but cannot do 2. This is not a complete victory for freedom, but it's a significant step forward.
Most of us do not live in a free-software utopia where we have direct control over every part of our machine from the firmware on up. (Although I have a lot of respect for the people who are trying to get us there!) Even if we're running a free OS, we assume, against our better judgment, that the firmware is doing only what it should be doing. If it's not, then it throws the rest of our freedom out the window.
Sandboxing EME allows us to ensure that one non-free part of the system, if we can't get rid of it, is restricted to doing only what it says it can do. This is a lot better than the status quo for any mixed free/non-free system (including free OS/non-free firmware). It's not quite the same as getting rid of it or making it free, but it's certainly not nothing.
A major issue is that arbitrary code can do arbitrary things, up to and including installing rootkits on your machine (for example, the Sony rootkit debacle). The sandbox means it can only do what the sandbox allows it to - in the case of Firefox's sandbox, that's very little.
"I should be able to watch whatever content I want on my system without installing proprietary software, even if that software is highly sandboxed and can't even see anything on my system and can't cause any permanent changes" is a political and moral issue more than a technical one.
Or use a free and open source browser that doesn't implement it. If this is something people want to take a principled stance against, those browsers will doubtlessly continue to exist.
Unless you're referring the content access. That's a general trend with technology that likely won't be stopped. If you choose not to use certain technologies, it's unsurprising that you won't be granted access to certain things.
I would like to see a security audit on that claim.
No saving means :
No time shifting, no device shifting, no legal fair use.
It does prevent these uses, however, unless the content provider explicitly provides it. The US law says that the content provider does not need to explicitly allow it; it's a right granted to the user by copyright law.
Cool, can you show me how to play the content on my portable MP4 player please.
Unfortunately, no. I think you may be getting tripped up by the phrase "effective technological measures" - being "effective" is a very, very low bar. To quote s296ZF of the Copyright, Designs and Patents Act 1988:[0]
“technological measures” are any technology, device or component which is designed, in the normal course of its operation, to protect a copyright work other than a computer program.
(2)Such measures are “effective” if the use of the work is controlled by the copyright owner through—
(a)an access control or protection process such as encryption, scrambling or other transformation of the work, or
(b)a copy control mechanism,
which achieves the intended protection.
[0] http://www.legislation.gov.uk/ukpga/1988/48/section/296ZF
Yeah, that's not true[1]. Sandbox security is not perfect.
Sandboxed plugins are definitely are an improvement in security over non-sandboxed ones, but running untrusted code in a sandbox is still running untrusted code on your machine. At the very least that has to be considered one level of privilege escalation that's just being handed to malicious coders for free.
Sandboxes are great. I'd love to see open source code run inside sandboxes, because even if the code has been audited, because even if I know the code isn't trying to do anything bad, it might have vulnerabilities that let other people do bad things. Running closed-source EME code inside a sandbox is still an unnecessary risk, though, because we have no way of verifying whether that code is secure and in fact that code might be an attacker itself.
[1] http://www.zdnet.com/article/pwn2own-2012-google-chrome-brow...
What I think is more important, from a security perspective, is that EME defeats (or attempts to defeat) attempts to get by without JavaScript or the complexity of modern browsers altogether. YouTube doesn't work as is without JavaScript anyway, but one determined not to use it could just download all their videos with youtube-dl; until someone comes out with a reverse engineered Widevine or Adobe Primetime or whatever other long-i-festooned DRM systems these sites support, this plan won't work for DRMed video.
asm.js and various obfuscators allow websites to do this in the first place with closed-source code, with a far larger attack service than the EME sandbox. In not a lot of time from now, they quite reasonably could.
I don't really think anyone has a problem with giving the latest episodes of Homeland and Downton Abbey to our political elite.
> You are absolutely deluded, if not stupid, if you think that a worldwide collection of software engineers who can't write operating systems or applications without security holes, can then turn around and suddenly write virtualization layers without security holes.
Net net, given the choice of extensive plug-ins vs. virtualization, I'll go with the latter.
That said, PPAPI is Chrome-only; EME is a standard, and Mozilla's implementation even outlines a way for other browsers to integrate the EME module Mozilla uses, plus their sandbox, into their own browsers.
Firefox also uses some kind of sandbox for plugins (like Flash) but yes, their planned sandbox for EME is stricter.
>> EME is a standard,
That's the heart of the problem. It guarantees the fact that web technologies are not open technologies anymore.
>> Mozilla's implementation
Mozilla dropped the ball years ago when it comes to proprietary techs. I see little difference between Firefox and Chrome on that matter.
Chrome and Firefox's sandboxes, meanwhile, are both open-source. You can inspect what powers the EME module might possibly have, and know that it can't gain any more. A vulnerability in the code is unlikely to be able to do anything other than pirate your download of Game of Thrones — and that's assuming it even has general-purpose network access. Ideally, a vulnerability would be able to do nothing other than modify the video you see, but the remote site could achieve that by encoding a modified video in the first place.
As far as the general moral arguments about DRM go, it's true that the new boss is the same as the old boss. But the bulk of the EFF's argument against Flash in this blog post is about security, not about open content, and it's important to acknowledge that EME is a significant step forward. The new boss is sitting in a tightly locked cage.