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

481–490 of 524 posts

Re: Update on HTML5 Video for Netflix

#481

Post by tptacek »

chii wrote: post #431
tptacek wrote: post #264

Earlier quoted context omitted.

EME strictly reduces the amount of "closed" code on the web. Without EME, more of the content protection stack is proprietary, and the trusted code base that enables content protection is larger. This is a very simple point that I think we all understand; I don't understand why EME's opposition thinks they can dodge it.

The assumption you make is that DRM _must_ exist in one form or another. I, as an open web proponent, want to make it difficult for any business to add DRM to their media offering. by making DRM a difficult system to implement due to proprietary plugins, it increases the perceived value of just offering it without DRM. but by having EME a standard, it makes it a no brainer to use DRM, because their customers will aut…

I have no trouble understanding the strategic goal nerds have of trying to use standards body formalities and browser vendors to retard the development of DRM software they disfavor. There's nothing wrong with that (at least, nothing more wrong with it than all the other things that are wrong with standards groups).

Where you lose me is the notion that there's an intrinsic ethical imperative not to provide attachment points in standards for DRM. That doesn't ring true.

Re: Update on HTML5 Video for Netflix

#482

Post by rbutcher »

braveo wrote: post #459
rbutcher wrote: post #451

Earlier quoted context omitted.

Netflix sends you a video that has been lossy compressed once. If you lossy compress that lossy compressed video, then you will have a video of worse quality than the video that was only lossy compressed once. Original source video > Netflix video stream (1 level of lossy encoding) > Recompressed video stream (2 levels of lossy encoding)

I think you'll find if you record and compress a netflix 4k video, the resulting quality will be just fine, especially if you're recording it and compressing it later (rather than doing it realtime). That's because, while it may have been compressed going over the wire, it's 4k and the quality is pretty stinkin' good. what you don't do is try to simply recompress the signal that netflix is sending you, you uncompress…

Nobody has said that the quality would be bad, just that some amount of generation loss [1] will occur when you use a lossily compressed [2] video (the netflix stream) as input for the creation of another lossily compressed video (the "recompressed" video file). The "recompressed" video would be slightly worse than the streamed video, which would be slightly worse than the original video.

An approximation of an approximation of a video is going to be worse than an approximation of a video.

1: https://en.wikipedia.org/wiki/Generation_loss

2: https://en.wikipedia.org/wiki/Lossy_compression

Re: Update on HTML5 Video for Netflix

#483

Post by cwyers »

k_ wrote: post #443
cwyers wrote: post #162

Earlier quoted context omitted.

I don't get it. You argue that not being able to visit Netflix "isn't the full experience," and so people won't choose a browser where they can't visit Netflix. So, because of this... all standards-compliant browsers should be unable to visit Netflix? So then why would anyone choose a standards-compliant browser?

It is not "not being able to visit Netflix" which isn't the full web experience. Web-based DRM will spread to other services, maybe even to other things than video. I couldn't care less about Netflix, I never intended to use their services and never will, but this really is bad news for the web. Netflix on the web with a plugin was fine for me: people wanting to watch Netflix on their standards-compliant browser woul…

This again seems odd. Browsers shipped with Flash preinstalled, and EME in Firefox is a plugin.

People are just tying themselves in knots to act like the status quo is a disaster.

Re: Update on HTML5 Video for Netflix

#484

Post by MBCook »

ailideex wrote: post #461
MBCook wrote: post #391

Earlier quoted context omitted.

Because a ton of people are honest and want to pay for content, and don't want to try that hard to pirate it. Even a small hurdle is enough to keep them on the straight and narrow. I'm quite willing to pay for content, I'll even go out of my way to find a way to do it sometimes. The few times I've looked into piracy in the last few years it's such a thorny mess that I'm quite happy to try and find a legal way to view…

IMO your missing the point - people wanting to pay for things legally does not somehow negate the existence of HDCP strippers and the pointlessness of DRM.

You're right, it doesn't. I just don't think most people, even if they know that exists, will go that far.

Re: Update on HTML5 Video for Netflix

#485

Post by criddell »

BuuQu9hu wrote: post #218
criddell wrote: post #163

Earlier quoted context omitted.

I'm still not getting how this is a step backwards. Last year there was no Netflix on Linux. This year there is. Is there a better way to make strong DRM?

The 'Open Web' means "My browser can access ANY AND ALL CONTENT on the Web given that it supports the necessary standards." This DRM implementation would add this suffix: "and any number of third parties have chosen to allow my browser to use their proprietary decryption modules." This is the opposite of the Open Web and therefore a threat to user freedom and the integrity of the Web as a whole.

The internet is used for a heck of a lot more than content destined for a browser.

Re: Update on HTML5 Video for Netflix

#486

Post by bagacrap »

always_good wrote: post #149
SwellJoe wrote: post #15

I don't like DRM, but I like binging Netflix in a corner on my laptop while I work. I've been doing it for years using Chrome with a fake user-agent to fool Netflix into thinking I'm on Windows (apparently that hasn't been needed for some time?) So...I'm torn, I guess. I'm glad Linux/Firefox can Netflix, but I'd like it if web standards were open and not patent encumbered. I guess it's a political issue and not a tec…

> I like binging Netflix in a corner on my > laptop while I work My girlfriend and some friends do this during work/chores, but I'm wondering how that works. Are there really shows that aren't good enough to warrant your undivided attention, yet good enough to watch in the first place? I really can't think of any for myself. If I don't mind missing chunks of a show, then I don't really want to watch it in the first p…

Werner Herzog documentaries for me.

Re: Update on HTML5 Video for Netflix

#487

Post by wildmusings »

problems wrote: post #417
Razengan wrote: post #365

Earlier quoted context omitted.

> content shouldn't be controlled (or that I, as a buyer of content, should control the content I bought) How should content creators protect against people who access their content without buying it?

They quite frankly can't technically, only legally - they're selling a 0 margin good, anyone who has it can basically give away infinite copies at zero cost per additional copy. Anyone who wants it for free will have it for free. They need no more than drop a few words into Google or type the name of something into a fully automatic tool like Sonarr. It's pointless for them to bother trying to stop piracy, especially…

"but it's not perfect, someone could just xyz and abc and get around the system."

DRM is not about being perfect against a dedicated pirate. Just like locks on doors, it's about putting up a few barriers that serve to keep the vast majority of people honest. No one is under any illusions that DRM will stop movies and games from showing up on torrent trackers.

Re: Update on HTML5 Video for Netflix

#488

Post by braveo »

rbutcher wrote: post #482
braveo wrote: post #459

Earlier quoted context omitted.

I think you'll find if you record and compress a netflix 4k video, the resulting quality will be just fine, especially if you're recording it and compressing it later (rather than doing it realtime). That's because, while it may have been compressed going over the wire, it's 4k and the quality is pretty stinkin' good. what you don't do is try to simply recompress the signal that netflix is sending you, you uncompress…

Nobody has said that the quality would be bad, just that some amount of generation loss [1] will occur when you use a lossily compressed [2] video (the netflix stream) as input for the creation of another lossily compressed video (the "recompressed" video file). The "recompressed" video would be slightly worse than the streamed video, which would be slightly worse than the original video. An approximation of an appro…

right, but what algesten originally said was (emphasis mine):

> _Problem is_ that grabbing the last frame buffer would mean you're doing it post decompression. To distribute something you grabbed like that you'd need to recompress it and then have a generation degradation of lossy compression.

In the context of getting a decent pirated copy, that's not really a problem due to how good those streams are.

That was my point, anyway.

Re: Update on HTML5 Video for Netflix

#489

Post by Silhouette »

flukus wrote: post #445
Silhouette wrote: post #440

Earlier quoted context omitted.

I'm not amazing at these things, but I could probably access a WiFi network unconnected to me, mask my IP to a different location, and acquire a random Netflix account in a couple of hours. Really? How would you do that, exactly? If you're looking online and find camcorder copies of videos that were served to a specific customer at a specific time, as confirmed by the watermarking, and your server logs show that that…

> Really? How would you do that, exactly? Wireless isn't very secure, you're mileage may vary on the encryption mechanism but here is a guide to hacking WPA networks (with WEP it takes seconds): http://www.hackingtutorials.org/wifi-hacking-tutorials/how-t... From there you can try a range of known exploits and gain admin access to a PC, after that it's game over, they can run what they want, when the want. There is n…

WPA2 is reasonably secure and has been the standard for home and business WiFi for years.

In any case, merely compromising WiFi won't get you someone's Netflix account. The Netflix data itself, including the credentials, are all encrypted.

As for taking over someone's PC, that's far beyond the average pirate, and you're talking about serious criminal offences on top of mere copyright infringement at that point.

And even then, we're not necessarily talking about being able to make exact copies of the stream. The whole point of hardware-backed DRM schemes is that just because you can run software on the PC, that doesn't mean you can access the unencrypted data stream.

So again, how exactly were you planning to do this? What you're talking about is far beyond the average script kiddie or casual pirate.

Remember, we were talking about convincing a court that it was more likely that someone did all of this and that was how a watermarked copy of protected content got out than that the person whose account was used to download that content then somehow shared it. A slight possibility that professional pirates who are also expert crackers chose that particular customer to pick on and left no evidence having done so isn't likely to be very convincing.

Re: Update on HTML5 Video for Netflix

#490

Post by kibwen »

rtpg wrote: post #441
kibwen wrote: post #161

Earlier quoted context omitted.

Mozilla did fight back, but nobody in the OSS community at large bothered to care about EME until the writing was already on the wall. I remember being at the Mozilla Summit 2013 discussing Mozilla's alternative proposals, but even by then the prevailing mood was that of imminent despair at the inevitable.

out of curiosity, what alternative proposals were offered?

AFAIR the alternative with the most promise involved implementing a new video codec in JS (heavily leveraging the GPU and the then-nascent asm.js) with first-class support for digital watermarking, then convincing studios that per-user watermarking was more effective than DRM. You can see the difficulty of Mozilla's position: it wasn't enough to come up with an alternative to EME, they had to come up with an alternative to DRM altogether. A tough sell, to say the least, and basically DOA because Chrome and IE were shipping EME for use with Netflix in 2012.
Post reply on HN