Mozilla's Route For Implementing W3C EME (HTML5 DRM)

[andreano]

I welcome HTML5 DRM

There are reasons to like this, from a blob hater's standpoint.

Think of what Flash & Silverlight already are. Just like a DRM-blob used in HTML5, they are Stallman's worst nightmare: unimplementable in free software, yet used as web standards.

Worse, creators & publishers don't percieve Flash & Silverlight as DRM, because they are so much more. Flash, at least, is being used recklessly, often where DRM isn't even intended. Consider bloggers posting videos of their cats ? they don't care for DRM and have no joy in pushing blobs on you. Despite HTML5 being super-easy for them to use, some braindamaged tool they like to use (is WordPress doing this?) wrap it in Flash, and thus you have this incompatibility layer (I call it DRM) between most web-video and your Raspberry Pi (or anything else not endorsed by Adobe). Totally unnecessary.

Silverlight, I doubt it is the minimal DRM-blob suitable for porting to lots of platforms, not to mention that maybe Microsoft would have a monopoly on porting it.

The industry wants DRM (again), this time they want to call DRM for what it is, and facilitate for minimal implementations and a competitive market.

[Luke]

If they don't control the kernel

Yeah, that was the scenario I was talking about, but implementation would seem to be important. For instance, the service could, periodically, recheck the binary during service (say, as part of the encrypted stream format), that's why I mentioned the, incredibly unlikely, possibility of finding a collision to exploit this method.
Is that a bit clearer?
Rerouting the process might be tricky since, iirc, web sockets are tied to the receiving process.

If they don't control the kernel they don't control and cannot trust the sockets, someone could patch it to deliberately allow socket exploits. Some think this sort of thing is the real reason for the locked bootloader/signed driver madness in Windows.

[smitty3268]

All major sites? Probably, they have lots of money and won't care paying for another license, might even have an Adobe license already due to Flash.

I don't know, what smaller sites will even be interested in DRM? I'd think they would largely be more likely to just go without it since it would cost them money to add DRM. I'm sure some porn sites will just grab MS DRM and force you to use IE to view anything, but sketchy sites are always going to be sketchy.

I don't know, maybe i'm naive, but i'm really hoping that this DRM crap will mostly be limited to Netflix, Hulu, BBC, and the like. Large sites that will want to hit everything. I don't think most small sites use DRM now in Flash.

[anda_skoa]

I don't know, what smaller sites will even be interested in DRM? I'd think they would largely be more likely to just go without it since it would cost them money to add DRM. I'm sure some porn sites will just grab MS DRM and force you to use IE to view anything, but sketchy sites are always going to be sketchy.

I don't know, maybe i'm naive, but i'm really hoping that this DRM crap will mostly be limited to Netflix, Hulu, BBC, and the like. Large sites that will want to hit everything. I don't think most small sites use DRM now in Flash.

Small site might have been the wrong term, maybe new or startup would have been better.
Bascially someone new to the scene who would want to compete, e.g. provide a service like Netflix in a country that is not served by Netflix.

I just don't see any value to license a fourth DRM just so that the few Firefox users on Linux who are strictly opposed to running Chrome can access the content. Those (and FirefoxOS users) are the only ones not covered by the platform providers' DRMs.

Cheers,
_

[liam]

If they don't control the kernel they don't control and cannot trust the sockets, someone could patch it to deliberately allow socket exploits. Some think this sort of thing is the real reason for the locked bootloader/signed driver madness in Windows.

OK, that's a possibility. I wonder if there is a cap that would provide guarantees for this?

[scjet]

EME, DRM, wow, how LOW can you go.?
So much for the "WWW", pathetic.

[Luke]

Big sites will also have all the badwidth

I don't know, what smaller sites will even be interested in DRM? I'd think they would largely be more likely to just go without it since it would cost them money to add DRM. I'm sure some porn sites will just grab MS DRM and force you to use IE to view anything, but sketchy sites are always going to be sketchy.

I don't know, maybe i'm naive, but i'm really hoping that this DRM crap will mostly be limited to Netflix, Hulu, BBC, and the like. Large sites that will want to hit everything. I don't think most small sites use DRM now in Flash.

In the US, those sites will both use DRM to limit use of their own videos, and take advantage of new FCC rules to buy up all the bandwidth to squash competing sources of video. I'm already considering distributing my end-of-the-year protest wrapup video on DVD at the end of this year.

[brosis]

There are reasons to like this, from a blob hater's standpoint.

Think of what Flash & Silverlight already are. Just like a DRM-blob used in HTML5, they are Stallman's worst nightmare: unimplementable in free software, yet used as web standards.

Worse, creators & publishers don't percieve Flash & Silverlight as DRM, because they are so much more. Flash, at least, is being used recklessly, often where DRM isn't even intended. Consider bloggers posting videos of their cats — they don't care for DRM and have no joy in pushing blobs on you. Despite HTML5 being super-easy for them to use, some braindamaged tool they like to use (is WordPress doing this?) wrap it in Flash, and thus you have this incompatibility layer (I call it DRM) between most web-video and your Raspberry Pi (or anything else not endorsed by Adobe). Totally unnecessary.

Silverlight, I doubt it is the minimal DRM-blob suitable for porting to lots of platforms, not to mention that maybe Microsoft would have a monopoly on porting it.

The industry wants DRM (again), this time they want to call DRM for what it is, and facilitate for minimal implementations and a competitive market.

Flash and silverlight started exactly like that.
DRM is very comparable to closed source driver. Its heavy limited to small amount of platforms.
There is no other way to implement a solution to legally use the content/card, unless DRM is cracked. This is why we have crappy nouveau performance and bugs, but radeon and intel drivers work very well - because industry opened up.
This is like things like S3TC in OpenGL open api. This is nothing, but a minetrap and has nothing to do in HTML standard.

Content delivery from A to B using some private keys is OKAY, but DRM is much much more and actually is a physical extension of legal ambition to CONTROL how content is consumed.

This is what you are missing from your thesis. If there is an industry that wants that, I will skip it altogether or support its cracking and making its status illegal.

[erendorn]

OK, that's a possibility. I wonder if there is a cap that would provide guarantees for this?

A cap?
In any case, if the user is in control of the platform (desktop windows and linux, rooted phones), binaries cannot be remotely trusted, while on platforms were the user is not (most portable OS including iOS and WP, gaming consoles, hardware players like boxes, etc..) then you can implement that. This can easily be observed in that there is much more cheating on PC games than on consoles, and that it is much easier to implement protected solutions (eg netflix) on portable devices and closed hardware (set-top box, console) than on PCs.

[highlandsun]

Mozilla should reverse this decision, and instead join the FSF in starting a class action lawsuit against Adobe and other purveyors of DRM.

erendorn's statement is right on the money:

DRM cannot rely on anything apart obfuscation. No amount of cryptography research will change this mathematical truth.

That obfuscation comes at a substantial runtime cost. E.g., the DRM that Adobe used in rtmpe used industry standard crypto algorithms, but heavily obfuscated. The obfuscated versions of these algorithms run thousands of times slower than a straightforward implementation, and consume hundreds of times more memory. DRM forces consumers to expend inordinate amounts of power and resources, wastefully. On this basis alone we have the seed of a case - the media providers are forcing consumers to pay double/triple/more for the same product, because in addition to the up-front purchase cost, there is excessive power cost each time the product is used.