Netflix on Linux with Chrome | Hacker News Reader

Netflix on Linux with Chrome

plus.google.com

131 pointscallumjones59 comments

In other news, bittorrent continues to work on all platforms, without placing any restrictions on what viewers can do.

Including participating in massive copyright infringement.

I struggle with the sentiment that if offered no other way to legally consume media, BitTorrent will show them! I can't find "The Money Pit" anywhere but BT. Does that make it ok to download?

I don't know. Morally, legally, etc., I've never been quite so torn for quite so long.

I think OP's point was that DRM pushes people to pirate content, regardless of whether this is morally acceptable or not.

Fundamentally, I don't think that DRM pushes people to piracy, I think poor user experience does. A lot of that has come out of poor DRM implementations, which is where the problem is. Spotify on the other hand, is based on a form of DRM but as they provide a good app for every platform I want to use (including Linux), and allow you to sync offline (for when I have no internet), I will quite happily give them my money and have no intention of pirating music ever again.

I think his point is rather that DRM asks more from the user (a specific browser, a specific plugin, a specific string to enter in a box) for essentially the same experience.

Oh and you can do less things with a DRM'd media than with a non-DRM'd. Even though you technically have the same rights.

Oh and you can do less things with a DRM'd media than with a non-DRM'd. Even though you technically have the same rights.

None of the big streaming-services supports the HW I have in my home-stereo rig. All of them supports H264 video and HDMI.

Not offering me H264 content uncrippled is leaving me, as a potential customer, no reason to sign up.

I sadly don't think that's true any more. Not that I want to promote piracy or am sad about the decline. But I fear that the upcoming generation of DRM will be good enough not to annoy the users. And that was really the issue with most DRM in the past. E.g., music is more commonly shared and copied to different devices. DRM there really had an impact. Flash/Silverlight are horrible plugins and they impact the viewing pleasure. But now that Restriction is integrated on top of the regular HTML5 stack in the top three proprietary browsers it will be as flawless as watching a HTML5 video for most users. Sure they can't save the video. But most users don't know how to do that even if the video doesn't have DRM. And as long as the business model of the content hosts is decent enough they won't feel bothered by the DRM. And unlike music where sharing and copying to different devices is what users expected long before MP3s, for videos the renting model is widely accepted and common since before computer video.

And that's one of the big issues with the EME proposal. It will ensure that DRM remains on video for a long time. It will make it impossible to view legal content in a complete free software stack. It will make the web depend on proprietary binary blobs from the three companies which are already controlling the majority web browser, operating system, and mobile device market share. Which means it will be hard to impossible to innovate in those areas if it's against the interests of those companies. This will have really damaging consequences.

It's a toughie because there are two evils at work here:

- DRM is its own kind of evil, but only implicitly because it erodes the rights and privacy of the users;

- Copyright infringement is another kind of evil, which is important because the law explicitly says so;

- Just not watching TV shows seems like the easiest option here, but it's boring and doesn't support starving artists.

Which kind of evil is worse?

> Copyright infringement is another kind of evil, which is important because the law explicitly says so;

It's not just the law that makes copyright infringement of various sorts immoral. It is the law, not morals, which gives GPL software strong copyleft, but surely you would agree that it would also be immoral to take GPL software proprietary without the permission of that software's authors?

You are comparing the act of taking a work and exploiting it for profit, and sharing a movie with a friend.

If try to commercial earn illicit profit by exploiting those people in society that gives software away for free and explicit allow you to run, modify and distribute. Then that is immoral. Comparing it to sharing a movie with a friend is just a shameful act of pushing an agenda.

A more correct comparison with GPL would be bootlegging a CD (ie, for profit) during a charity concert that has pay-what-you-want tickets. Anti-copyright people would likely support the claim that such behavior is immoral.

> Anti-copyright people would likely support the claim that such behavior is immoral.

Yes, that is my point.

But GPL works in practice because the immoral behavior it is meant to defend against is also illegal due to copyright.

The original point was that copyrights merely make things illegal, but that we shouldn't care because what is illegal would be something moral (I mean, is there an easier target to beat up on than RIAA and MPAA?).

I countered that in this case there are instances where copyright makes immoral behavior illegal—it is not simply true to state that any activity illegal under the Copyright Act would be moral to engage in, if not for those pesky IP Nazis oppressing the public.

Copyright infringement is another kind of evil, which is important because the law explicitly says so;

But the intent of copyright law was that society as a whole should be enriched.

Given how the law has been corrupted into a prison for cultural content and a right for the rich to profit, while society as a whole gets poorer and poorer arts-wise, I find it morally just to encourage civil disobedience.

Firefox + pipelight + UA switcher has been working flawlessly* for me for the last six months or so.

It doesn't work on chrome anymore unfortunately after the update where they disabled NPAPI plugins.

Native support still can't come soon enough.

i don't think the explanation as to why this works is correct. the poster says it's because Google and Microsoft "have been working closely with Netflix to get the needed DRM code required for HTML5 streaming", yet:

1. the user-agent switch involves adding a Safari identifier (rather than IE or a new Chrome one)

2. at WWDC 2014, Apple announced that it had been working closely with Netflix along those lines, and that the upcoming version of Safari would be able to play Netflix without Silverlight. (And indeed Safari on Yosemite can do just that in the developer previews)

That is a chrome UA, it has Chrome/X in it. Safari/X is a compatibility shim.

That is the Chrome user agent. It mentions Safari and AppleWebkit for compatibility reasons but it's Chrome.

Windows Chrome user-agent for reference:

  Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/36.0.1985.125 Safari/537.36

Ah, yes, another reminder of the hilariously bad user-agent situation in today's web...

(And indeed Safari on Yosemite can do just that in the developer previews)

Yep, it's definitely a bit buggy and stops occasionally but it's usable!

"the user-agent switch involves adding a Safari identifier"

Well, the default user agent string for Chrome already has a Safari identifier in it, but I don't think Safari user agents have a Chrome identifier. The change being made in the post doesn't add a Safari identifier, because there was one there already, but it does change the OS string to Windows. So I think the explanation in the post is probably right.

For a humorous history of why exactly the Chrome UA string claims to be Safari, see this:

http://webaim.org/blog/user-agent-string-history/

It's not complete without the new Windows Phone user agent, though:

https://mobile.twitter.com/0xabad1dea/status/497122542885556...

I thought the whole point of "HTML5 DRM" was to be backed by hardware DRM. So are they giving up on that idea for Linux? And if they are, maybe hardware DRM isn't necessary on other platforms either.

HTML5 DRM (EME) only defines an interface between the DRM system and the browser, it doesn't specify the mechanism for the DRM itself, the CDM is a black box . You could have a DRM scheme based on ROT13 and it would be entirely valid.

https://dvcs.w3.org/hg/html-media/raw-file/tip/encrypted-med...

What they mean is that the "Digital Restriction Management in HTML5" proposal is pushed by Google, Microsoft, and Netflix into the W3C spec. Apple has simply followed the proposal and implemented it as well. AFAIK Chrome, IE, Safari ship with it. Firefox will follow suite by adding a Restriction Module from Adobe (given Adobe's track record it will certainly be an amazingly fast, flexible, beautiful, efficient, and nice way of viewing videos...).

So Web-Restrictions is kinda unstoppable now. Instead of having a plugin-free future with HTML5 we have an even worse situation.

"No DRM" was never an option. The option was between 1 DRM standard and 50 (so silverlight and widevine and....).

The W3C made the most pragmatic choice, here. I think DRM is still crap, but one system is unambiguously better than >1.

Except Flash was at least used for stuff other than just DRM, so there was an incentive to get it working across all OSes. Linux is going to struggle once this is adopted, at least open browsers will, and that makes me sad

The openness of the browser has nothing to do with it, the browser provides an interface to the DRM mechanism which could be open or closed source. Open source browsers like firefox already support proprietary plugins such as Flash.

If anything this makes Linux support easier, because you no longer have to provide support for an entire virtual machine and all APIs such as Flash or Silverlight. The only software required will be the DRM module itself as everything else can be provided by the browser.

The DRM mechanism can't really be open source. You could simply change it to write the unencrypted video data to the disk if it were.

A plugin DRM module will also have to do all of the video rendering and displaying. Because if it just hands the unencrypted stream back to the browser then you could change the browser to simply write it to disk.

The relevant DRM mechanisms will come from Apple, Microsoft, Google, and Adobe. Both Apple and Microsoft have no interest in supporting GNU/Linux. You won't get their DRM on GNU/Linux no matter how simple it is. Google seems to ship their DRM now on the GNU/Linux version of Chrome. But I don't know if the license allows using it in other browsers. Which would be hindered by the fact that Google uses their own unspecified PAPPI. Adobe wants to provide a DRM module for Firefox EME mechanism. But we all know from enough bad experience how well Adobe does GNU/Linux support. Flash on GNU/Linux was even worse than on any other system until they simply stopped it. They recently even stopped distributing Acrobat Reader for GNU/Linux. So yeah, great hope there. Even if they compile the module for GNU/Linux because Mozilla asks them to then we can expect the typical Adobe software safety and quality... Remember HTML5 was supposed to rid the world of Flash, Silverlight, and such things. Not force those binary blobs into the spec.

Having an EME spec allows for competition in the DRM market, it creates a business opportunity for a company (or potentially a solo dev) to develop a cross platform solution.

EME means you rely less on Adobe code than you did previously as they are no longer shipping an entire runtime, just the content protection module.

You are completely wrong here. Please note that the EME proposal does _not_ specify a plugin interface for the Restriction module! It only specifies how the Restriction module is exposed to JavaScript. How the Restriction module is implemented or connected to the browser is up to the browser developer. Both Google and Microsoft are involved in the creation of the spec and Apple is also supporting it. All three companies make up a large share of the web browser market. And all three companies have their existing DRM solutions which they are using for their EME implementation. None of them have announced a plugin interface to allow other companies to provide a DRM module.

The only browser vendor who wants to implement EME via a plugin is Mozilla. Simply because there can't be a free software Restriction module implementation (and consequently due to the W3C's efforts there defacto can't be a fully free software web implementation). And Mozilla already made a deal with Adobe to provide the Restriction module.

So no, there won't be a market. It will actually close down the market. Content providers will have to support those four DRM solutions if they want to offer their content on all those platforms. They won't have any choice.

That browser vendors will bundle their own solutions doesn't really close down the market and I don't understand how it would necessarily hurt cross platform adoption?

Of course it closes down the market. There won't even be a market. There is not going to be a plugin interface (except for Firefox) to provide a different Restriction module.

It's not just 1 DRM standard, it's 1 simple DRM standard that is only DRM.

The DRM plugin is basically just request/provide license and decrypt frame. Even if it was encrypted media extensions vs silverlight that's 100k plugin vs 10 MiB plugin.

Actually, it's 1 DRM _plugin_ standard. There are still 50 DRM standards. Safari ships with Apple's DRM, Chrome ships with Google's DRM (Widevine), IE ships with Microsoft's DRM, Firefox doesn't ship with any DRM (but its EME plugin is only compatible with Adobe DRM).

> Instead of having a plugin-free future with HTML5 we have an even worse situation.

Indeed. Now we need to have one plugin, per browser, per operating system supported.

Guess which next emerging OS wont be able to fully access "standard-compliant" HTML5? Yeah. All of them.

W3C should be thrown to the sharks for this.

Blaming the W3C is less effective than blaming the specific W3C members, and browser vendors, who had the power to push this through. Specifically: Google and Microsoft.

Those two companies formulated EME together with Netflix.

I think blaming the W3C is absolutely correct here as well. The W3C has a principle to defend: The open web. And they are simply throwing it overboard by allowing this. TimBL even made a statement in favour of it. Yes, refusing to accept the standard won't change the reality because Google, Microsoft, and Apple are in favour of DRM and they'll do it no matter what the W3C will say. But at least the W3C would stay true to the principles of an open web.

The most offensive part about this whole shenanigans is the sheer hypocrisy behind it.

Hollywood and its friend likes to claim that they need DRM to prevent piracy, to prevent freeloaders from getting use of something they shouldn't. Looking at the big picture though, they are the freeloaders and they are shameless about it too.

They are more than eager to freeload on all the good parts about digital media (zero cost distribution, zero cost storage, zero cost reproduction, etc) while they insist that the business around it should still be limited to the ones they had with tangible, physical media and goods.

Not only that, they want to freeload on all the open, accessible, cross-platform technology we have developed and built to sell their restricted content. And they have the balls to demand that these open standard (which has been successful because of their openness), should now be closed to accommodate their needs!

They are hypocrites from end to end. They want and want and want, but are not willing to give an inch. They demand that we give up our freedom and valuable standards. And for what? To be able to see their content on a restricted service while we pretend its not available in a superior DRM-free format illegally everywhere else on the internet.

It's just disgusting.

Why not just let the whole DRM thing go? Let us access the content in a superior, portable format and don't ruin the future of the web over the semantics of how it gets projected onto the customers screen of choice, on their platform of choice.

If you use Ubuntu 14.04 (like me) you have to install a newer version of libnss3 (at least 3.16.2) and than it works.

Runs smoother than pipelight + firefox at least for me.

Full instructions for reference. Run:

  wget http://us.archive.ubuntu.com/ubuntu/pool/main/n/nss/libnss3{_3.16.3-1ubuntu1_amd64,-1d_3.16.3-1ubuntu1_amd64,-nssdb_3.16.3-1ubuntu1_all}.deb
  sudo dpkg -i libnss3*.deb

Create a new Tampermonkey script:

  // ==UserScript==
  // @name         Netflix HTML5 for Linux
  // @namespace    0d0a3443-cc79-4f86-bf62-2dac581d2b3a
  // @version      0.1
  // @description  Enables HTML5 video support for the Linux host platform
  // @match        *://www.netflix.com/WiPlayer?*
  // @user-agent   Mozilla/5.0 (Windows NT 6.3; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/37.0.2062.68 Safari/537.36
  // ==/UserScript==

I tried this, but I just get prompted to install the silverlight plugin. I am using Lubuntu 14.04.

In your account settings, try turning on the HTML5 Player. Its opt-in.

CentOS-6.5 and later with Pipelight and Firefox 24.7.0 ESR works flawlessly. God knows what it's installing and what security holes I opened up by downloading it though.

Looking forward to any native solutions.

I can confirm this works on Chrome on OpenSuSE 13.1:

    Version 37.0.2062.68 beta (64-bit)

Really, it's more like accidently supported on Linux.

But I thought that Netflix required Silverlight on the client to work. What is this using then?

The HTML5 encrypted media extensions. Chrome has support for that

Doesn't work for me. Error Code: M7063-1913

lsb_release -d: Ubuntu 14.04.1 LTS

uname -a: Linux x230 3.13.0-32-generic #57-Ubuntu SMP Tue Jul 15 03:51:08 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux

chrome://version: 38.0.2114.2 (Official Build 287444) dev

Does this work with only Chrome and not Chromium?

Yes. Only chrome includes the Widevine binary blob and supporting source code in the browser.

Chromium is open source doesn't ship with the proprietary, closed bits of Chrome, like the EME DRM module which is used here. So this is Chrome-only.

Open-source software can obviously never properly support DRM.

Which is why Mozilla/Firefox was so much against this whole HTML5 DRM bullshit in the first place, and so should you have been to.

If anyone gets it working, my start-up is doing some interesting things with Netfix: turboflix.com

Your website seems confusing a few ways:

* You only list TV shows in the examples, despite the service offering both movies and shows

* I actually do have all of those shows (or at least most) on my Netflix. Although I see on your Facebook page that the service is perhaps targeted to people in other countries.)

* I am wondering if what your service offers will be legal, especially if you are charging for it (my best guess is that it is some kind of proxy that lets you access international Netflixes, perhaps a little more convenient than existing Chrome extensions in that it might list movies allowed in different countries together.)

Fantastic.

What do you use to watch movies and series on-demand ?

Why, a web browser !

It works perfectly on v37 on my Mac OSX machine (running Mavericks)

Anyone tried it on a Raspberry Pi?

google-chrome not supported on arm

Say what? The majority of Android devices have ARM processors, and Chrome is available for Android:

https://play.google.com/store/apps/details?id=com.android.ch...

Reply on news.ycombinator.com