Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Update on HTML5 Video for Netflix

techblog.netflix.com

Post reply on HN
Print view View original

271–280 of 524 posts

Re: Update on HTML5 Video for Netflix

#271

Post by caf »

Ironchefpython wrote: post #244
mavrc wrote: post #230

Earlier quoted context omitted.

> And if they did and the industry said no, what should they have done? Distribute industry content with DRM, distribute their own content without, and see what the results are. Somebody's going to have to perform the experiment to see what the outcome will be. Since Netflix produces tons of content now, they are in an excellent position to try this.

> Distribute industry content with DRM, distribute their own content without, and see what the results are. The industry content will show up on the Pirate Bay, and Netflix original content will show up on the Pirate Bay. Were you expecting another outcome?

..which would indicate that the marginal value delivered by the DRM is zero?

Re: Update on HTML5 Video for Netflix

#272

Post by bubblethink »

mschuster91 wrote: post #200
TeMPOraL wrote: post #152

Earlier quoted context omitted.

Did it? I have a feeling we're about to see more DRM with music. All the pieces are falling into place. Files get replaced with streaming (which is to files what a service is to a product), and I wonder how long it will take before more companies will follow Apple's suit by ditching the audio jack. Going full digital there will enable companies to set up fully DRMed path from memory to speakers/headphones, like they…

> to set up fully DRMed path from memory to speakers/headphones Unlike with video, where the signal ends up unencrypted only on the very last chip in the chain, namely the one that drives the (I'm taking a guess here!) thousands of pixel lines, this is not practical with audio signals - you can always intercept and easily record the signal in analog form, either right before the PA stage or if this does not work (bec…

The battle isn't about recording the decoded stream. Even with video, all forms of HDCP have been broken. It's about getting it before it's decoded.

Re: Update on HTML5 Video for Netflix

#273

Post by Nexxxeh »

nkristoffersen wrote: post #219
ghughes wrote: post #84

Earlier quoted context omitted.

Nope, there's a company in Shenzhen that makes a plug-and-play bypass device: https://arstechnica.com/tech-policy/2016/01/warner-bros-sues...

I used the hdfury years ago. the output quality wasn't anywhere near the quality of the input signal. but it did technically give you a signal without hdcp.

Different generations, and presumably you were going DAC, to VGA, rather than staying digital.

Re: Update on HTML5 Video for Netflix

#274

Post by BuuQu9hu »

tptacek wrote: post #265
BuuQu9hu wrote: post #261

Earlier quoted context omitted.

This is false. EME describes only the interface between the Javascript environment and the CDM. It says nothing about the CDM's interface with the browser code, which is the heart of the issue.

You're rebutting an argument I didn't make. The CDM is closed-source, but it was closed-source in the pre-EME world too. My argument is that EME opens up DRM stacks; it's that it opens up more of the DRM stack than was open before it.

> by standardizing everything but (in essence) the authenticated key derivation step.

Your argument is that EME covers most of the code responsible implementing DRM. This is false. EME is a small part. The large majority of the implementation is not standardized. For example, there is no equivalent of NPAPI in the post-EME world; the CDM/browser interface is browser-specific.

Re: Update on HTML5 Video for Netflix

#275

Post by problems »

pornel wrote: post #202
problems wrote: post #156

So I remember reading some early EME specs that seemed to indicate that EME encrypted content could be decoded and played by the browser's standard decoder and player - in the case of an open source browser like Firefox - hooking these decoding functions is quite trivial. Is this the means they've gone with or does this EME blob include their own decoder and player? EDIT: This little image here: https://www.w3.org/TR…

This diagram and most of non-normative parts of the spec are a deliberate red herring. The purpose of these is to derail discussion about DRM restrictions by saying that any restrictions are optional. However, during discussions at W3C Netflix representatives confirmed that their implementation doesn't work like this. They are only interested in "strong" DRM model in which the player is not trusted and has no ability…

On Android they feed it directly to the OS codecs except for 4K video so they can't really care that much.

Re: Update on HTML5 Video for Netflix

#276

Post by e40 »

jameskegel wrote: post #270
e40 wrote: post #256

Earlier quoted context omitted.

I don't agree with "people want to watch what they want to watch." I have quite a few data points that say you are flat wrong about that. I would add to this: people watch what they want to watch and how they want to watch it. I've literally downloaded HBO shows even though I'm an HBO Now subscriber. The Roku HBO Now app is so horrible, I just hate using it.

I agree with this. My frustration is not only HBO, but also UFC. I pay for Fight Pass because I love the sport, and I want to support it, but my Roku experience is so bad that I found myself looking for other means of consuming it.

Me, too (on the UFC)! I never use it. I just pay them because I download the content elsewhere.

Re: Update on HTML5 Video for Netflix

#277

Post by pikzen »

tptacek wrote: post #101
josteink wrote: post #55

Earlier quoted context omitted.

> Netflix wouldn't be a viable business if not for DRM. Why is everyone so upset? Having the option to use DRM in native web APIs is not a bad thing. Because Netflix is freeloading on our open technology to deliver closed content, and putting the cost to support their DRM on us. It's not right. It's not fair. They are fragmenting the web by making "web sites" which are browser and OS-specific. If they want to make cl…

In what sense is Netflix "freeloading"? Where does the free ride stop? Are they allowed to use TCP/IP stacks to deliver DRM? The HTTP protocol? Or is just Mozilla that's the problem?

You don't see in what sense forcing the implementation of DRM into an open standard (HTTP in a broad sense, which includes HTML, JS, etc.), that has been developed for the past decades in a rather open way so your company can keep making profits could be "freeloading"? Not to mention setting it up in such a way that their part of the web is broken if for some reason you do not have their blessing for their DRM.

Netflix came in and screwed over everyone, in a way that is just as bad as Flash.

Sure, use the TCP/IP stack. Send your data over the internet. Make your own software based on that, it's just a transport layer and it screws no one else. Except they seem to be pretty damn awful at it, considering how comically bad all their desktop apps are.

Re: Update on HTML5 Video for Netflix

#278

Post by caf »

greenhouse_gas wrote: post #132
jasonmp85 wrote: post #3

Earlier quoted context omitted.

Yes, the world was better when Flash and Silverlight were the media of choice for delivering video.

The thing is that there is (for all intents and purposes) no difference between ECE and plugins. Actually, ECE is a plugin. The only difference is that it helps web-designers switch plugin providers. That's it. No more security. No more standardization. As a user, you still depend on the goodwill of the ECE provider throw you a bone. If you use a slightly niche OS (say, OpenBSD on ARM), you're out of luck. If Adobe d…

Google's CDM ("Widevine") does run on Firefox, apparently.

I wonder whether it would be feasible to run the CDM within QEMU on a platform like OpenBSD/ARM? At the end of the day it's just ordinary code that runs in the Gecko Media Plugin sandbox.

Re: Update on HTML5 Video for Netflix

#279

Post by coldtea »

ReverseCold wrote: post #246
dasil003 wrote: post #231

Earlier quoted context omitted.

Sorry, but as someone who spent nearly a decade founding and building a premium content VOD service, and 6 of those years fighting fiercely against DRM and making great content available without DRM, I have both the knowledge and moral authority to say that you're living in a fantasy world. You might as well say that anyone who is not living an RMS-sanctioned ascetic technology lifestyle is hurting the cause of free…

Someone can fork FF though and make it so that the DRM looks like it's there to the websites it visits, but it isn't really there. There's no way DRM can't be removed if I own the device that can play it.

>Someone can fork FF though and make it so that the DRM looks like it's there to the websites it visits, but it isn't really there.

No, they can't. The DRM will have to be there, or the content (as is from the website) wont play.

>There's no way DRM can't be removed if I own the device that can play it.

There are several ways (including recording the analog signal if it comes down to that). But not any way of "make it so that the DRM looks like it's there to the websites it visits, but it isn't really there"

Re: Update on HTML5 Video for Netflix

#280

Post by zzalpha »

dasil003 wrote: post #231
cyborgx7 wrote: post #2

Translation: Netflix successfully lobied DRM into what is supposed to be the OPEN web standard, to the point where even Firefox couldn't afford not to support it. Dark days for the free exchange of information enabled by the web.

Sorry, but as someone who spent nearly a decade founding and building a premium content VOD service, and 6 of those years fighting fiercely against DRM and making great content available without DRM, I have both the knowledge and moral authority to say that you're living in a fantasy world. You might as well say that anyone who is not living an RMS-sanctioned ascetic technology lifestyle is hurting the cause of free…

tl;dr Content is king.

And you're absolutely right.

Post reply on HN