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

VP8 and H.264 to both become mandatory for WebRTC

andreasgal.com

Post reply on HN
Print view View original

71–78 of 78 posts

Re: VP8 and H.264 to both become mandatory for WebRTC

#71

Post by eridius »

shmerl wrote: post #68
eridius wrote: post #57

Earlier quoted context omitted.

I can't decide if you genuinely believe that Apple is deciding not to spend resources supporting FLAC because you think they have a "nasty attitude", or if you're deliberately misrepresenting the reasons that lead to such a decision. Supporting FLAC requires investing engineer resources in doing this, and possibly legal resources as well. It's only something that Apple would do if there's any benefit to them doing it…

> Supporting FLAC requires investing engineer resources in doing this, and possibly legal resources as well. It's only something that Apple would do if there's any benefit to them doing it. FLAC is patent free and actively used (commercially including) by many parties big and small, so this legal FUD is totally unconvincing. ALAC isn't supported in hardware any better than FLAC, so that argument completely misses the…

> And for over than a decade "couldn't find resources" to support FLAC which is actually used unlike ALAC.

Are you really this uninformed, or do you just like to make things up when the facts don't go your way?

ALAC was created for the purpose of streaming audio between Apple devices. I believe it was first used to stream audio to an AirPort Express (which has audio output, so you can plug a speaker into it and stream music from iTunes to that speaker). I assume it's still used for that purpose, but it's also used for the more general category of streaming audio over AirPlay. Given that, I would wager that ALAC is used many orders of magnitude more than FLAC is, even if it's not directly exposed to the end user.

Re: VP8 and H.264 to both become mandatory for WebRTC

#72

Post by shmerl »

eridius wrote: post #69
shmerl wrote: post #66

Earlier quoted context omitted.

> That assumes their users actually ever encounter such music. Not "their users", any users. FLAC is the only lossless format that's being actively used commercially (and not commercially) by various services and stores. So, their users encounter it as well when they deal with lossless audio. > Anyone who gets FLAC music is choosing to do so Yep, since it's the only practical lossless format offered as above. And App…

A cynic would point out that anyone who is purchasing FLAC music is, by definition, not purchasing from the iTunes Store, so there is very little business reason to facilitate this behavior. A more practical person would say that nearly everyone who does purchase from an alternative store doesn't want FLAC anyway (or at least, shouldn't want it, though it would not surprise me to see a lot of people who opt for FLAC…

> For those vanishingly small number of people who have an actual use for FLAC

This approach of dumbifying users to brainless consumers of content is extremely annoying. I consider any company that treats all users that way to be simply insulting. I don't mean those who cater for non technical users and being very user friendly. I welcome that. I mean those who proclaim that they can cripple functionality of what already exists with their justification that all users are dumb and would never need it.

For instance, audio CDs offer two features - good quality of sound and lossless data (i.e. which you can reencode to any lossy format with transparency without degrading quality). Those features are there for years already. Now, comes the digital age and normal services offer a substitute - FLAC. Comes Apple and says - users are dumb, no one needs that functionality. No CDs for you, and no lossless audio either. Take our AAC, and if you want to reencode - get lost.

Well, that's insulting. But Apple aren't alone in it. For instance you can't buy FLAC on Amazon either. On the other hand, other services respect their users more and give an option of lossless audio. Just because they can and it's easy to provide. See Cdbaby, Bandcamd and etc. They offer FLAC files along with a variety of lossy codecs. But this isn't even so much about sellers. We are talking about support in players / systems. One should be a jerk not to provide support when one can treat users with more respect and simply add that support for most common lossless codec.

> assertions that Apple is intentionally choosing not to support FLAC out of some personified desire to screw their users is ridiculous, right?

I see no valid reason for them not to support it, when their own users asked them multiple times to do it. Apple dismissed them. So they do want to screw their users. Or may be simply their dislike of free codecs is stronger than their interest to help their own users. I honestly see no single valid reason for them not to do it, especially since it's trivial. It's available in every possible third party player imaginable. But Apple? No, they pretend it doesn't exist.

Re: VP8 and H.264 to both become mandatory for WebRTC

#73

Post by shmerl »

eridius wrote: post #71
shmerl wrote: post #68

Earlier quoted context omitted.

> Supporting FLAC requires investing engineer resources in doing this, and possibly legal resources as well. It's only something that Apple would do if there's any benefit to them doing it. FLAC is patent free and actively used (commercially including) by many parties big and small, so this legal FUD is totally unconvincing. ALAC isn't supported in hardware any better than FLAC, so that argument completely misses the…

> And for over than a decade "couldn't find resources" to support FLAC which is actually used unlike ALAC. Are you really this uninformed, or do you just like to make things up when the facts don't go your way? ALAC was created for the purpose of streaming audio between Apple devices. I believe it was first used to stream audio to an AirPort Express (which has audio output, so you can plug a speaker into it and strea…

ALAC was pure NIH, since FLAC existed before it. And it wasn't even open until much later. Anyway, it's not any excuse for Apple not to support FLAC. It's irrelevant to this subject really.

Re: VP8 and H.264 to both become mandatory for WebRTC

#74

Post by shmerl »

eridius wrote: post #70
shmerl wrote: post #67

Earlier quoted context omitted.

> True. But this is an issue for so few people as to be effectively zero. Why not? Keeping a master copy can be an issue for any user who cares about quality. You can see it as keeping a master tape, so any subsequent copy (=lossy encoding) won't degrade the quality too far. > and I would wager that most of them aren't Apple customers to begin with. Why so? Is it some kind of stereotype that Apple customers don't car…

> You can see it as keeping a master tape, so any subsequent copy (=lossy encoding) won't degrade the quality too far. Yes, the concept is not what's at question here. What's at question is how many people actually are inclined to ever care about something like this, and the answer is almost nobody. Very few people care to reencode their audio files. > Why so? Is it some kind of stereotype that Apple customers don't…

> Then there's the ongoing maintenance burden, of dealing with upstream changes, local bugfixes, and just plain integration with the rest of the QuickTime stack.

Yes, and Apple perfectly did all that when it came to their own ALAC, as you already mentioned. Not FLAC though. NIH I guess. Or just "screw your customer". Such serious company like Apple being unable to support FLAC with QA and upstream updates? I don't believe that. They just don't want to. And not because it's costly (it's nothing for them), and not because they are scared of legal threats (there are none - it's used in tons of places just fine). It's just Apple being Apple the way I see it.

Re: VP8 and H.264 to both become mandatory for WebRTC

#75

Post by anon1385 »

shmerl wrote: post #52
anon1385 wrote: post #33

Earlier quoted context omitted.

>Nothing stops Mozilla from supporting JPEG2000 except their nasty attitude in general. JPEG2000 is actually supported by many image editors, unlike APNG. I think in most of these cases the real reasons are more mundane - spending the resources on supporting extra formats would give them no competitive advantage (and a significant cost in terms of maintenance and security). It sucks, but that's the capitalist system…

> Nothing stops Mozilla from supporting JPEG2000 Patents stop them (JPEG2000 is not a free format). FLAC is not patented. Q.E.D. Apple don't supported it just in order to be nasty to everyone.

>Patents stop them (JPEG2000 is not a free format).

Oh please, stop with this FUD. JPEG2000 is no different to Ogg or VP8 in that regard, both of which Mozilla is happy to ship. It was designed to be usable without having to licence any patents. Any known patents have been waived. The possibility of unknown patents remains but is vanishingly small by this point (other organisations much larger than Mozilla have been shipping JPEG2000 code for years).

http://www.jpeg.org/jpeg2000/index.html

>Furthermore, it includes guidelines and examples, a bibliography of technical references, and a list of companies from whom patent statements have been received by ISO. JPEG 2000 was developed with the intention that Part 1 could be implemented without the payment of licence fees or royalties, and identified patent holders have waived their rights toward this end. However, the JPEG committee cannot make a formal guarantee, and it remains the responsibility of the implementer to ensure that no patents are infringed.

What the fuck does Mozilla even mean by "free format" these days? It's an ISO standard, it was designed to be usable without paying any fees, there is a waiver of any known patents, it's been used across the industry for years without legal problems. I can't think of a single way that it could be made "more free". The only possible reason I can see is that it wasn't invented by Mozilla/Xiph. And really that is just pathetic and very much a "nasty attitude".

Re: VP8 and H.264 to both become mandatory for WebRTC

#76

Post by nitrogen »

eridius wrote: post #65
shmerl wrote: post #63

Earlier quoted context omitted.

> Supported everywhere != used everywhere. In my experience, extremely few people use FLAC, Way more than ALAC. Music in FLAC is provided by many music services and digital stores. Music in ALAC? I never saw it being sold anywhere. > because there's almost never a reason to care about having a lossless audio codec. That's utter nonsense. Any time you want to reencode your music, you care about the lossless codec for…

> Music in ALAC? I never saw it being sold anywhere. Because almost nobody has any reason to want lossless music. Anyone buying music in FLAC is deluding themselves if they think they can hear a difference between that and a properly encoded lossy codec like MP3 or AAC. > Any time you want to reencode your music, you care about the lossless codec for the source, otherwise you'll degrade your quality. True. But this i…

Anyone buying music in FLAC is deluding themselves if they think they can hear a difference between that and a properly encoded lossy codec like MP3

Because of inadequacies of the MP3 encoding format, no bitrate of MP3, even the max, can encode all possible things that the human ear can distinguish. There is one song I know of with a particular synthesized effect in the upper treble that is very profoundly different from the lossless original even in a VBR0 or 320kbit MP3.

Additionally, I've read that MP3 (I don't know about AAC) doesn't preserve enough phase information for effective use with matrix-encoded (e.g. Dolby Pro Logic) surround sound audio.

But Apple's engineering resources are not infinite, and it would be a flagrant waste of those resources to spend them on issues like this that impact effectively zero Apple users.

What was wasteful was inventing ALAC instead of using FLAC.

Re: VP8 and H.264 to both become mandatory for WebRTC

#77

Post by shmerl »

anon1385 wrote: post #75
shmerl wrote: post #52

Earlier quoted context omitted.

> Nothing stops Mozilla from supporting JPEG2000 Patents stop them (JPEG2000 is not a free format). FLAC is not patented. Q.E.D. Apple don't supported it just in order to be nasty to everyone.

>Patents stop them (JPEG2000 is not a free format). Oh please, stop with this FUD. JPEG2000 is no different to Ogg or VP8 in that regard, both of which Mozilla is happy to ship. It was designed to be usable without having to licence any patents. Any known patents have been waived. The possibility of unknown patents remains but is vanishingly small by this point (other organisations much larger than Mozilla have been…

It's not fully patent free:

> It has always been a strong goal of the JPEG committee that its standards should be implementable in their baseline form without payment of royalty and license fees...

It's only for baseline.

> is no different to Ogg or VP8 in that regard

They are patent free, no strings attached, unlike JPEG2000.

Re: VP8 and H.264 to both become mandatory for WebRTC

#78

Post by ZeroGravitas »

derf_ wrote: post #24
shmerl wrote: post #23

I hope Daala will put an end to this mess. But even though Opus is mandatory now, it didn't yet translate into support by Apple and MS for instance for regular music and Web audio. Their historic sickening opposition to open codecs is not easy to dismantle. Apple still doesn't even support FLAC, just because they like to make things messy for everyone. By the way, what happened with Nokia's attacks on VP8? Were they…

Cisco hired a law firm to write a summary, which they made public: http://www.duanemorris.com/memo/VP8Compilation.pdf TL/DR: VP8 was found not to infringe one patent. The other proceeding in the German courts was stayed until it was determined if the patent was valid. Before that could happen, Nokia and HTC settled all of their outstanding suits for some undisclosed amount. All remaining actions between those two par…

Nokia just announced an Android tablet, running Lollipop 5.0 (which requires VP8 and VP9), it also uses an intel chipset with VP8 encode and decode.

Does that trigger any of the reciprocal patent grant thingies that VP8 (and 9?) have and remove Nokia's patents from the picture?

Post reply on HN