Slashdot Mirror
The Ambiguity of "Open" and VP8 Vs. H.264
An anonymous reader writes "With all the talk about WebM and H.264, how the move might be a step backwards for openness, and Google's intention to add 'plugins' for IE9 and Safari to support WebM, this article attempts to clear misconceptions about the VP8 and H.264 codecs and how browsers render video. Firefox, Opera and Google rely on their own media frameworks to decode video, whereas IE9 and Safari will hand over video processing to the operating system (Windows Media Player or QuickTime), the need for the web to establish a baseline codec for encoding videos, and how the Flash player is proprietary, but implementation and usage remain royalty free."
Please make it easier to report/flag spammer accounts. That is all.
vos nescitis quicquam, nec cogitatis quia expedit nobis ut unus moriatur homo pro populo et non tota gens pereat.
H.264s development was open? I mean really that is just a bit of a reach.
Far more so than VP8's development was until last May. At least with H.264 it was being developed between different companies and industry groups whereas VP8 was a closed-source, proprietary codec developed by a two-bit company that almost no consumer before Google's buy out had every heard of.
The only thing that concerns me about the web video format is that it needs to be unencumbered by royalties or other licensing. If I want to make a video, encode it, sell it, make ads off of a website, get 100 or 100,000 visitors, I should damn well be able to do that without having to pay a dime to anyone for the ability to make my own god damn videos--unless I optionally choose.
By using h.264, you pretty much guarantee that *someone* *somewhere* is paying for it. Could you imagine if say, the "David After Dentist" kid had to pay tons and tons of royalties to the MPAA for a video they created simply because they used the h.264 container format? To even conceive such a thing is such bullshit that this should absolutely be a non-issue.
Though this will never happen, the US government should claim eminent domain on all patents involving the h.264 technology, and then dare the large companies to make a move. After all, we're the ones with the guns.
H.264 is closed. VP8 is open.
How is H.264 closed? The spec is available for any one to buy and implement. If H.264 is "closed" than so can be said for the vast majority of ISO standards.
It's frustrating that only the OS-provided solutions (Safari and IE) are doing this right by handing it off to the OS. The notion that your browser needs to reimplement everything, including video rendering, is what leads to the bloatware we have today. The whole point of having an OS is to have a common framework and API layer that all applications hosted on it can access. Instead, Firefox, Chrome and Opera are all re-developing their own video rendering, for each platform they exist on, AND each one needs to write its own video-card accelerator layers for each platform it exists on.
I'm out of my mind right now, but feel free to leave a message.....
Unfortunately, that means you don't get H.264 on Linux as its a proprietary codec that requires some form of (paid) licencing.
I mean, Firefox doesn't support H.264, but Microsoft will happily provide you with the capability of playing H.264 in firefox using a driver that leverages the OS capability... as long as you're running it on Windows.
I think you're partly right though, all the codecs should be implemented as drivers (or similar) and then you are technically using the OS-provided capability, once the correct codec is installed. But its not like the OS is providing the drivers directly, you'll haver to go get them from somewhere. As WebM is free, codecs for it will be freely available for all OSs.
I guess the problem comes for those OSs that are locked down, but then you'er always on to a loser - if Apple only supports H.264 on iPhone and Microsoft only supports (say) H.265 on WP7, and neither allows you to upgrade the video support, then you will never get a video to play universally.
At least there's no excuse for not supporting WebM by all manufacturers, and any who try to give one will quickly be found out by consumers.
As an analogy - look at the non-free 'internets', Microsoft tried to lock you into MSN, and AOL tried similarly. Look where they are now.
If H.264 is "closed" than so can be said for the vast majority of ISO standards.
Not sure if it's a vast majority, but a lot of ISO standards are closed. Even so closed that you cannot read them without paying a shitload of money.
Really? Can you contribute code to H.264? Can you use the spec in your own software and publish it with out a large amount of jumping through hoops?
Really H.264 may have been public but I would not call it open. WebM is now what I would consider to be open as is Theora and Dirac http://diracvideo.org/ .
So no I do not feel that H.254 meets the definition of open as far as development goes.
So yes it really is a bit of a reach IMHO.
See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
There are open standards, and open source, and they are not the same. The IETF, for example (subject to yesterdays Birthday Article) deals with open standards. Linux, by contrast, is open source.
An open standard means that no one party controls the generation of the standard, and that the standard is openly available. Generally, open standards are developed by SDOs (Standards Defining Organizations, such as the IETF or the W3C). As a general rule "anyone" can participate in their creation (but this may require that you or your company be a member of some organization or have some other qualifications). Many open standards have patent encumbrances. Typically, SDOs seek RAND (Reasonable and NonDiscriminatory) licensing terms; some even require a particular patent licensing policy as a condition for participation. The IETF, however, requires disclosure and seeks, but does not strictly require, RAND terms. While an open standard may have some code associated with it, typically the entire point of an open standard is to allow you to go off and write your own code, generally under whatever code license you want. This is how the Internet was developed.
Open source means that the source is licensed by GPL or BSD> or some similar licensing. Now, generally open source means that the code is available, but in practice many open source projects are more or less closed to outside participation, and they frequently do not provide documentation sufficient to replicate what they are doing.
H.264 is an open standard, governed by standards bodies (ISO & ITU). It is not, however, a 'free' standard, in either the "beer" or "freedom" sense.
VP8 is a proprietary standard, governed by Google, and developed by a single company. It is, allegedly, a 'free' standard in both the beer and freedom sense - and it's worth noting that there are some concerns as to whether or not this standard would survive an IP infringement claim, making it less "free" than we're asked to assume.
You're right, the definitions are quite clear. I'm just not sure why you seem to think it's opposite day when labelling H.264 as closed and VP8 as open. Until Google submits VP8 to ISO or some other standards body, it's not an "open" standard, it's a "Google says it's cool so I guess that's what we should do," standard. It would seem that you're conflating "royalty-free" with "open."
Really?
Yes, really. Before Google opened the code in May of last year, On2 was developing VP8 as a closed-source proprietary codec since 2008. H.264 on the other hand was developed by the ISO standards board and a whole host of companies in it's development. Like all ISO standards one could get access to the full spec. Such a thing was impossible for the first 2.5 years of VP8's life.
Really H.264 may have been public but I would not call it open.
Can you use the spec in your own software and publish it with out a large amount of jumping through hoops?
Sure, x264 developers have been doing so for the better part of 6 years.
It's no less open than most of the other standards which are called "open".
So no I do not feel that H.254 meets the definition of open as far as development goes.
And neither was VP8 until 7 months ago when it was a completely closed-source codec.
"And neither was VP8 until 7 months ago when it was a completely closed-source codec."
Well then this post would have been right 7 months ago. But that was seven months ago and this is now.
See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
So the Right Thing is to force everyone to buy an OS from Microsoft or Apple? Do you know there are some crazy people developing free operating systems? And even using them! How dare they ask for a royalty free baseline codec for encoding video for the web?
You're missing what the GP said - no-one's suggesting forcing anyone to buy an OS, the suggestion is to hand off video playback to the OS. In this case, the right thing to do would be to release it to a video decoding layer for Linux and then call it from Firefox/Chrome.
Cheers,
Ian
And that's why I said:
Far more so than VP8's development was until last May
Secondly, H.264 is no more "closed" than the supposed "open" standards such as ISO C++ with statements like:
Can you contribute code to H.264?
To turn it around, can YOU contribute to the C++ ISO standard? Highly unlikely just like it's highly unlikely that most people could contribute to the H.264 ISO standard. So by this logic C++ is also a "closed" standard, no?
OpenDocument Specification: http://www.oasis-open.org/committees/tc_home.php?wg_abbrev=office#odf11
I don't think it stands up to any of the FOSS definitions of "open".
And the same could be said about the C++ and ODF standards yet those are called "open" standards by the same people talking about how H.264 is "closed".
VP8 - maybe it wasn't open 7 months ago, but it is now.
Is it really? Can any individual really have any meaningful say in the direction of how the VP8 codec is developed unless you work at Google? Sure they've given the source out but you'll have no more say in how the spec develops than you would for the H.264 standard.
A good point here - Google has a lot of "green" initiatives (reduced-power computing, huge solar cell farms on their roof, etc.)
This approach is NOT a "green" approach - a "green" approach is one that makes use of the large amount of hardware acceleration infrastructure now deployed for the existing standard codecs.
WebM/VP8 will force a non-accelerated CPU-only rendering path on ALL existing hardware. This eats power compared to hardware acceleration. (Look at how well most Android devices handle H.264 thanks to hardware accelerated decoding.)
Google is being hypocritical and inconsistent here. Great summary at http://daringfireball.net/2011/01/simple_questions - Key here is, HTML5 was supposed to at least partially break Adobe's stranglehold on the web by moving some content away from Flash. Google just killed any hope of that - They talk about supporting open codecs, but they still bundle Adobe Flash (which includes H.264 support) with Chrome?
As a result of this mess, content providers are starting to shy away from HTML5 and stick with what "just works" (for the most part) - SmugMug was starting to consider HTML5, but Google's latest decision has them moving back to Flash.
retrorocket.o not found, launch anyway?
Can you contribute code to H.264?
The question does not make sense. It's like asking 'can you contribute code to HTML?' H.264 is a standard, not an implementation. The license of various implementations is independent of the way in which the standard was developed.
H.264 was developed jointly the ITU-T Video Coding Experts Group (VCEG) and the ISO/IEC Moving Picture Experts Group (MPEG). These groups solicited contributions from anyone. If you wanted to contribute something to the spec, you could. There was a lot of political stuff as well, with a few things being added to the spec just so that companies could get one of their patents in.
In contrast, VP8 was developed in private by On2 and dumped on the public by Google. The x.264 developers raised some issues with the spec, but were told that the format was frozen and would not be modified. Theora and Dirac are both frozen now, but they had an open development process and modified the bitstream format several times based on feedback from external groups.
So, when you are talking about the process for developing the spec, Theora, Dirac, and H.264 were all open. When you are talking about using the spec, Theora, Dirac, and VP8 are all open.
I am TheRaven on Soylent News
In this case, the right thing to do would be to release it to a video decoding layer for Linux
Which would end up supporting only MPEG-1, Theora, and VP8 given the patent policies of many GNU/Linux distributors. And for each operating system, how should the browser direct the user to find and install appropriate codecs? Do video decoding layers for Linux even support codecs installed by one user for that user as opposed to codecs installed by root for all users? Most of the tutorials I found were for .deb installation on Ubuntu, which is always system-wide.
Thank you. Someone finally understands what I'm saying. The problem is that so many other standards that work in the exact same way that H.264 did are referred to as "open" yet H.264 is demonized as being "closed" despite there being little to no difference in the way both standards were developed.
But that is in the past, by focusing on it now, you are making it look like (in fact making the argument) that H.264 is more open, through focus on and old irrelevant fact, but ignoring another definition of the word open where WebM is much more open than H.264 will ever be.
Let's take this:
* According to one aspect H.264 was once more open, but this aspect applies to the past.
* According to another aspect WebM is much more open, and this applies today.
I am not saying you are wrong, you are in fact right, but you are distorting the debate through pedantic and irrelevant details.
Now you didn't start this doublespeak, but I can only think the person who did, was either doing so deliberately or is in serious denial.
Even if you had a time machine, you still couldn't contribute anything to the VP8 "standard" - it was developed entirely by that single company, and now the bitstream has been fixed and google are not accepting improvements or even obvious bugfixes. Wheras h264 was a real ISO standard - everyone was welcome to speak (though of course not necessarily be listened to) in the standardization discussions, and every country got to vote.
I am trolling
Secondly, even if you are distrbuting binary encoders/decoders you don't pay anything until you hit about 50,000 units shipped.
This is the problem with x264. If x264 becomes the de facto standard, two guys in a garage will never be able to develop their own browser that competes with all the current market leaders, because the second it starts to gain widespread acceptance it becomes subject to royalty fees that two guys in a garage will never be able to afford. The x264 standard may be open, but you can't do anything useful with that standard without paying up.
We are talking about now, though. I agree that H.264 is an open standard and VP8 was a closed one, but WebM is an open standard now and this is what should really matter at this point.
The critical difference between the two formats now is that one is royalty free and one is temporarily royalty free - in other words, we have no idea how H.264 could evolve. Maybe it'll stay royalty free forever, which would make it an interesting alternative. Maybe it will not, though, and that could be a potential disaster for video on the web - or just a thorn in the side of Google and other big video sites.
The big debate therefore is: do we stay with a widely adopted, high performance format that may behave like a Damocles sword, or do we switch now for what is currently an inferior but safer alternative?
The C++ standard I can make a compiler for without paying anyone. It is not a burden to entry like h.264 is.
The ISO stopped meaning anything the minute they approved the MS "open" formats.
He was being rhetorical when he asked "really?".
I think h.264 has done a great job for the web. It's provided us with high quality video on demand. It's helped ensure our hardware also has high quality video.
VP8 on the other hand, regardless of its' roots is meant to help break a lock on the industry, a lock that h.264 has gained. It's a lock that must be broken. Having choice is really all that matters even if it sets things back once in a while. Often times industries take 2 steps forward and 1 step back.
Technically, this is not a huge change. It isn't an instant change. If the industry can implement this in the web and other software products, as well as hardware, then so be it. If both need to be supported then so be it. It's not unheard of and not altogether uncommon.
The goal is to give choice and to ensure that the consumer isn't locked into one product, that, in being so, denies them choice and increases their costs.
So, so be it. Nothing we do here in debate will change the reality of the situation. Google's made a choice that it feels is best to ensure that things are open and inexpensive.
Time to move forward.
You can lead a man with reason but you can't make him think.
I would say the difference between them is patent encumbrance. Sure you can use h.264 if you're a smelly basement dwelling open source fanatic, but commercial usage is limited by patent licensing and royalties.
Why are you lying?
FFMPeg is GPL
x264 is also GPL
Do I need to go on and list a few more, or is two enough to snub your ignorance?
He's not lying, he's just over-simplfying.
So far, software patents have not been legally applied to source code because source code has been clearly defined as "speech" as it is a means for people to express ideas.
So it is legal to write and distribute source code.
But, in most countries with software patents, it is illegal to actually use a binary built from that source code.
Its just the compiling it yourself or downloading it from a country without software patents makes it pretty much impossible to get caught.
When information is power, privacy is freedom.
You are confusing the standards with their implementations.
All of these standards are now frozen, so no one can contribute to them. H.264 was open during its design, and VP8 was closed (and suggestions for improvement were ignored when the spec and reference implementation was made available). Since they are both frozen, I'd say H.264 spec was and is more open *as a standard*.
Now, as far as implementations go, it's a different story (though still not as cut and dried as people claim). VP8/WebM is now open source, great And x264 is a GPL implementation of H.264, so it is just as "open". The difference all comes down to licensing - a number of patents are required to implement the H.264 standard, so anyone who implements it and wants to use it in a country that recognizes those patents has to pay licensing fees or risk being sued.
That last bit definitely makes VP8 more attractive to people who don't want to pay license fees. So, call it "more expensive to use", "patent encumbered", or some other more descriptive term. But just throwing around the vague concept of "open" without the real context doesn't help the discussion...
- AVC/H.264 FAQ
"National Security is the chief cause of national insecurity." - Celine's First Law
The problem, of course, is we don't know whether VP8 will stay royalty free either with the patent threats hanging over it.
Which specific patent threats? I'm not talking bullshit random "there might be a patent threat somewhere hiding under the wardrobe" patent threats. I'm talking threats with a patent number and a "you are infringing, pay up or else" letter attached to them.
Let me make a patent "threat". There might be a secret H.264 patent that which I might have heard of which which will maybe suddenly come to life next year. If you don't pay me a million Euros for every device you have I might not use my (possibly existing or possibly not existing) influence to divert this threat that may (or may not) appear later.
Anybody can do that. If you fail to specifically notify someone who has put a public implementation out for free, what they have done wrong you aren't fulfilling your duties as a patent holder wanting to collect royalties.
And with Google refusing to indemnify users of the spec, and refusing to take legal action to get a legal opinion (from a court - what are those called?) that it violates no patents, one can't be sure whether MPEG-LA's rumbling has any basis in fact.
Strangely enough the MPEG-LA also provides no indemnification and has failed to "legal action to get a legal opinion". What Google provides, for free, is a license for all patents known to be used in the WebM standard, exactly the same as the MPEG-LA charges for.
What is interesting is; what is the source for your ideas? Where did you even get the idea that Google is "refusing to take legal action"? It's impossible to prove a negative and it's impossible to take action against widespread innuenduo. No judge will grant an open statement that "no patents are infringed". At best they could act to say "patent number XYZ was not infringed. You should look over that source agan and see if it's not trying to mislead you over a bunch of other things.
=~ s,(.*),<sarcasm>$1</sarcasm>,g if any_point_you_wish();
But with WebM VS H.264 everyone seems to be happily ignoring the elephant in the room...WebM sucks. I'm sorry, but it does. You are pretty much gonna have to toss every mobile device or watch the battery go dead so fast you might as well just drag a very long extension cord with you and call that mobile. H.264 hardware acceleration is built into just about EVERYTHING, desktops, laptops, pads, phones, TVs, PMPs, etc. Are we gonna throw every single one of those devices out now? Wow that will do wonders for the environment.
The facts are these: Even if everyone started work right this second you are probably looking at around 2 years before WebM support trickles down to the chips in our everyday devices. And of course every device already out built in the last four years or so, plus everything coming down the pipe for the 2 years it will take to add support for WebM? They will all run H.264 great and WebM like ass. So I really hate to burst Google's bubble but it's a little too little, a little too late when it comes to WebM. Like Vorbis by the time it got to the party the world had decided on something else (in that case a likewise encumbered MP3) but you know what? It is only geeks and FOSS advocates that give a flying crap about "free as in freedom man, yeah!" and the rest of the world just throws a few pennies to MPEG-LA or whomever and doesn't give a fuck.
So if Google wants to pull a MSFT and shoot itself in the foot, go right ahead! it still won't make flash or H.264 go anywhere, as a matter of fact I predict this will further entrench flash as one will only have to leave a "raw" H.264 file for iDevice users and drop the same file in a flash container for everybody else, thus making their lives easier. What Google is attempting is to make another tower of Babble like we had with RMV VS WMV VS QT back in the 90s, but the common man just isn't gonna put up with that shit anymore. If Youtube doesn't work on their iDevice or it sucks their new Droid deader than Dixie when they use it? Well screw Youtube then, it isn't like there aren't a bazillion sites where they can see funny cat videos out there and they sure as hell ain't giving up their new iShiny toys for stupid Youtube clips, now are they?
ACs don't waste your time replying, your posts are never seen by me.
Yet, oddly, it is the smelly basement dwelling open source fanatics who are complaining most about H.264. The others out there who really have a product to sell realize the licensing fees are really minimal.
Patents also encumber USB and HDMI, I haven't seen Google on the barricades against those technologies used in Android phones and Google TVs respectively.
Over 20 hardware manufacturers are working on WebM hardware implementations, including Broadcom and Qualcomm, the two biggest chipset makers for mobile devices. When H.264 was standardized, all computer implementations were done in software as well. The hardware acceleration came later. Three years ago, HD-DVD and BluRay war was still undecided, and smartphones that played streaming video all but non-existent. Who knows how much inroads WebM could make in the next three years.
SmugMug was starting to consider HTML5, but Google's latest decision has them moving back to Flash.
Firefox and Opera don't support H.264 either, and they have much greater market share than Google. So if this announcement changes anyone's plans, they obviously hadn't thought them through very well to begin with. Either you support two formats for the next several years until everything is sorted out, or you exclude a large portion of your audience. This is a draft standard we are talking about. You should expect early adopter issues.
This is the problem with x264. If x264 becomes the de facto standard, two guys in a garage will never be able to develop their own browser that competes with all the current market leaders
Any browser writer could implement the video tag in such a way they fall back to system supported codecs. Then they need not pay anyone even though on all platforms you would support h.264 playback.
"There is more worth loving than we have strength to love." - Brian Jay Stanley