A Blog From Behind the Trenches

I think they can also use quicktime as a plug in for h.264

As pointed out on a tweet yesterday I saw h.264 can't be the w3c standard for html5 video

It seems lots of people woke up yesterday and found the internet. People defending h.264 just proves their ignorance on the matter.

Originally posted by xErath:

It seems lots of people woke up yesterday and found the internet. People defending h.264 just proves their ignorance on the matter.

How do you figure? (not trying to be a wiseguy, I really don't grasp the gravity of this situation)

Originally posted by ernesteban:

How do you figure? (not trying to be a wiseguy, I really don't grasp the gravity of this situation)

Did you read the blog post ?

maybe Google should offer a free version of Chrome with WebM+Theora and a paid one with +H.264, I'm pretty sure Peter Bright wouldn't be willing to pay for the later (especially as web video := YouTube := Google := WebM)

The comparison of Flash still being bundled even though H.264 is to be removed was ridiculous to me. Thats where I stopped reading his article.

Exactly, "Flash is already ubiquitous." The advantage of bundling Flash at this point is that, since its already used by so many people and has been for a long time, Google can now make sure that they are at least using the latest version for stability and security (via auto-updating the plugin).

What if tomorrow MPEG-LA announce that h.264 is "royalty-free"? Something like this...

h.264 and the codecs it supports (video and audio) require no royalty payments of any kind. You can do whatever you want with the h.264 code without owing money to anybody.

Is there any chance that Google will change his decision in favor of a superior h.264?

Originally posted by InDigazzZ:

What if tomorrow MPEG-LA announce that h.264 is "royalty-free"?

That's not going to happen. The MPEG-LA are seeking revenue from everyone in that food chain, from people who produce video content and use it commercially (Read the EULA/License agreement for a video camera that outputs h.264), to the people who produce encoders, those who publish content. The only exception is the production/distribution of video online without a fee.

http://www.businesswire.com/news/home/20100825006629/en

Originally posted by InDigazzZ:

Is there any chance that Google will change his decision in favor of a superior h.264?

Probably, they will be forced to. But the MPEG-LA would never do this, even when it's codec won't be the web standard there will be plenty other media where they can earn royalties.

Originally posted by InDigazzZ:

What if tomorrow MPEG-LA announce that h.264 is "royalty-free"? Something like this...

h.264 and the codecs it supports (video and audio) require no royalty payments of any kind. You can do whatever you want with the h.264 code without owing money to anybody.

That could happen, in fact MPEG-LA have already announced something similar. However, saying "h.264 requires no royalty payments" is VERY VERY different to saying "h.264 will not require royalty payments".

It is very easy for a product to add a feature that is temporarily royalty free, it is something different altogether to attempt to remove that product when royalties are brought in and the format is ubiquitous. Google are taking the wise move of removing h.264 now, when HTML5 video is still budding - if they don't remove it now, it would be impossible for them to remove later if h.264 in conjunction with the video tag became hugely popular on the web, and MPEG-LA could at that point charge whatever they see fit.

Originally posted by InDigazzZ:

Is there any chance that Google will change his decision in favor of a superior h.264?

H264 isn't superior to WebM. It has some advantages, but using the word "superior" is not quite right.

It's also interesting how Peter Bright basically recommends that Firefox and Opera should just distribute H.264 codecs illegally. He doesn't offer any solution himself or calls out Microsoft or Apple to finance H.264 as web codec, if they like it that much.

Obviously the GIF example also went largely over his head. He assumes that the GIF software patents applied to browsers back then (in reality only the LZW compression, not decompression demanded royalty fees). And the 256 color myth makes the whole rant appear thin on the technical topics.

So, can we expect Google to drop mp3 from Chrome for

In 6 months time, of the HTML5 capable browsers, Firefox, Opera, Chrome, IE9, Safari, more 2/3rds (by usage stats) will support WebM exclusively and less than 1/3rd will support H.264 exclusively.

H.264 is not winning when it comes to HTML5 video support. WebM is winning. Actually, it's more than just winning; it's kicking ass.

"H.264 can be used in both Flash and through HTML5 video, ensuring a smooth transition"

In addition to your points, Flash is getting WebM support soon (source: http://blogs.adobe.com/flashplatform/2010/05/adobe_support_for_vp8.html ), so this argument isn't really valid.

Originally posted by mcritz:

So, can we expect Google to drop mp3 from Chrome for

Not a valid argument since it's already got a huge market share, there's not much that can be done. H.264 however can still be removed from the picture.

And furthermore, I'm pretty sure chrome doesn't actually ship a mp3 decoder, but rather uses the operating system's native mp3 decoder.

Originally posted by haavard:

2. "VP8 is not an open standard"

That is actually correct.

Hey, thanks; I wasn't aware of that.

Finally some sense on the subject, great post

I, personally, often use H.264. Offline.
And I'm happy, that Google tries to give us a better (free) choice. I'm not FLOSS maniac freak, but some standards and core technologies just have to be free and open.

Originally posted by mcritz:

So, can we expect Google to drop mp3 from Chrome for

Maybe. But video is much more important, and the outcome is much less certain. Mp3 is already everywhere, like Flash..

Originally posted by AsaDotzler:

In 6 months time, of the HTML5 capable browsers, Firefox, Opera, Chrome, IE9, Safari, more 2/3rds (by usage stats) will support WebM exclusively and less than 1/3rd will support H.264 exclusively.

It's easy to add WebM support to Safari and IE as well. Just install the codec on the system, and they will automatically pick it up and use it. So all you need to do is to create a small installer to quickly and easily install WebM as a system codec, and the user is good to go.

Here is my problem with decision. In my job creating websites for our clients my mandate is to create sites that work on the widest range of platforms and browsers. This includes Safari on iOS devices. Currently this means I have to encode videos at least 2 different ways - h.264 and .ogv. I still have to wrap video in flash for other browsers, and lately I've been encoding in .flv and .webm as well. This is pretty much a pain in the ass. Google dropping support for h.264 does exactly nothing to my need for h.264'encoded video -- I still need it to serve iPhone, iPad, iPod, other cellphone makes, and Safari users. Is Google actually telling the public to dump billions of dollars of equipment? Are you? I'm curious as to what your recommendation would be for producers of web content such as myself who can't really afford to turn some customers away.

Originally posted by JamesHT:

This is pretty much a pain in the ass. Google dropping support for h.264 does exactly nothing to my need for h.264'encoded video -- I still need it to serve iPhone, iPad, iPod, other cellphone makes, and Safari users.

One could just as easily say "Apple not supporting WebM does nothing to my need for WebM-encoded video -- I still need it to serve Chrome, Firefox, Opera, other browsers." You're simply making a choice about who you point the finger at.

And in this particular case, the choice you're making is "do I blame Apple for my encoding workload, or do I blame EVERYONE except Apple"...

For those saying that h264 isn't widespread. Please be aware that it is
-The video format of choice for most large video sites, including YouTube, Netflix and Vimeo
-The defacto format for video podcasts
-Part of Blu-Ray (granted that format years are numbered)
-Supported by billions of devices

For those ready to flame me, please understand Chrome is my favorite browser by far, and I'm very much for open standards, I'm the Web Director for my company, so making the process more streamlined is very important to me. H264 because of the broad support noted above did that. My main issue is that I'm also very much against re-encoding 1000's of videos for our company and clients. We went down the h264 path because it is the current standard, not to mention the backing of Microsoft, Apple and Google at the time. My main concern is the time and resources its going to take to rework solutions that work better than WebM can currently or probably will for a year or more to come.

What I don't get is, and I could be wrong, of course, but because Google must license h264 for YouTube they've likely payed the full license cap (my understand is its $6.5mil), so they could easily/legally distribute it. Open standards part I get, but broadest reach I don't. Will they drop it from Android or Youtube next... doubtful.

There are a number of problems with many of your points:

1. This is accurate until you state: "By the very definition of 'open'..." You mean to use the word "free," not "open." Yes, the w3c's Patent Policy prevents its inclusion because it is not free, but this does permit being sloppy with our use of terms. H.264 is an open standard. It is not free. This debate is too easily derailed by the more ideologically extreme to use "open" as "open and free" when "free" suffices for "free."

2. WebM is not an open standard, and it would go along way for those hesitant to use WebM if Google would submit it to any one of several standards body for open review and participation. The lack of such action is quite troubling to me. Google wants it to be a standard but doesn't want a standards organization to review it. Why? Because Google permits a reciprocal license without fees doesn't make it a "good candidate," it makes it a qualified candidate.

3. I agree that H.264 is disqualified for inclusion in the HTML5 specification because currently necessary licenses require payment. I reject that this is necessarily a bait-and-switch scheme. I have more faith in market forces, legal requirements, and competition applying reasonable pressure on the consortium of patent holders than I do in conspiracies.

4. This one set off red lights and alarm bells: "But it would become another closed de facto standard, just like IE6. And we all remember the damage that did to the web."

Firstly, let's address the bogey man of IE6. Yes, late nineties behavior by Microsoft was not enjoyable, but I cannot accept history repeating itself in the present. I also remember that, as a Mac user, I switched to IE because Netscape had a worse product, and I switched to Gecko/Firefox because it was better, outside of the competitive pressures Microsoft applied to Windows. I now use Chrome and Safari. We are in a more competitive environment with more players, with the EU willing to rapidly investigate anticompetitive behavior, and new players like Facebook emerging. Not having faith in the OSS community, Adobe, Google, Apple, Microsoft, or an individual or small company to compete with H.264 is disheartening.

But let's get back to the actual point: "But it would become another closed de facto standard." Really, how does that happen? By Google continuing to include optional native support of a codec, H.264 "becomes a closed de facto standard"? Even though they still embed (not provide a plug-in) Flash which provides native support of H.264 and plug-ins can be created to add it to the browser or use a System codec? Huh, that's magical. Or is it that H.264 is already a de facto standard?

In virtually every digital non-web market it certainly already is and has been heading there for eight years with a rapid hockey stick over the last two. Even on the web, it makes up the majority of the encoded video. All of the major competitors in web video (Adobe - Flash; Microsoft - Silverlight/IE; Apple - everywhere; Google - YouTube, Android, Chrome; as well as the myriad alternatives to YouTube) are supporting and using it in major ways in their web strategies despite fundamental disagreements. Every web-based service not interfaced via html has assuredly and without controversy standardized on H.264. They rely on it, support it, use it, encourage it, and agree that it is the best solution. Nearly every device, software, and service supports it already. The CFP for H.264 was issued in 1998, formerly approved as an international standard in 2003, hardware-accelerations began to appear in the months after, and has 8 years of adoption, psychovisual optimizations, and improvements to hardware-acceleration which continue today.

But Google including it as an option in a browser (that can easily gain H.264 via a plug-in anyway) MAKES it become a de facto standard rather than these realities?

5. "This is comparing apples and oranges. Flash is a plugin, which Google chose to bundle because there is a lot of Flash content out there. On the other hand, H.264 would be part of the browser itself, and not a mere plugin."

The distinction you are making does not matter to me as someone who understands and uses the technology, and certainly doesn't register to the average end-user. Not only is the distinction irrelevant semantics ("native" v. "plug-in"), but the manor in which Flash is now integrated with Chrome 5 certainly disqualifies any rhetorical distinction: it can still be turned on and off as a plug-in and is sandboxed, but I certainly hear Google and Adobe describing it as "native" Flash support. The distinction is only relevant to those who fear the w3c specification will matter less than the real web (sorry, it always will), and who, for some reason, feel that "native" versus "plug-in", "embedded", or "system-enabled" is one slip too far down a slope we're already moving down at a rapid pace.

As to ubiquity and need, see above ("de facto") and Flash NEEDS H.264 to provide video support for the web today. If Flash is needed, so is H.264. Preferring a problematic plug-in integrated as native code into an application solely to provide access to a codec, despite that plug-in also carrying the baggage of a meta-platform runtime, is not preferable to direct access to the NEEDED codec.

6. "Just because a format is widespread offline does not mean that it is suitable for use on the web." This is an agreeable statement on first inspection. Once you realize that the video must created first -- off the web -- it's illogical to introduce a web-only format. It's not merely widespread, it is THE CHOICE everywhere offline. The idea that it's practical that everything on the web exist as B format when it is created, stored, edited, produced, distributed, and consumed in format A does not seem sensible merely to "prop up" free. It is already getting paid for everywhere else! The idea that the web is a wall that requires we submit to Google, apparently (via YouTube), all the world's H.264 video to be reprocessed back out as FREE to cross to the other side of sanctified seems far less desirable than a commercial codec that is being used everywhere else anyway.

"Since the web requires open standards, H.264 is not suitable as the primary format for video on the web, by definition." No, it makes it not suitable as a candidate to the w3c specification which guarantees the baseline accessibility of the web. My web has never been just the w3c specification. No browser has ever been just the specification. We cannot equate "the web" with the w3c specification. The w3c specification determines the best available baseline for compatibility and accessibility for the web; it does not circumscribe the entirety of the web. It simply has never worked that way and never will. I encounter proprietary every moment I am on the web.

The web is, above non-proprietary, practical, which is why there is much that is proprietary or non-specified throughout it. To have an offline format and an online format is not practical. It's actually quite absurd.

7. Again, The Great Re-Encoder in the Sky argument. Rather than make the reality of the situation easy to use on the web; the entire offline world submits all digital video to be reprocessed and stored and delivered in another format... on the web! ...again, using Google to do so! ...using their format! ...for the sake of freedom! Providing a baseline "free" video format for the web is fine; denying it a better, more useful, more practical video format that can be used universally without re-encoding is not fine.

8. The fact of the matter is that although Apple and Microsoft have made their codec format decision, both actually allow access to system-installed codec options. It is the "open" browsers (excepting Opera which can utilize system codecs) choosing to eliminate native support or system codecs as a choice. However, they too can easily use plug-ins. Little has been accomplished in attempting to deny access to H.264; it only makes it less easy, useful, and practical. We should not be pushing towards plug-ins or maintaining Flash's ubiquity.

9. "Google's online advertising monopoly is working on overdrive to ensure that won't happen." Google is an impressive corporation, but this is your argument for how WebM eclipses H.264? A single company buying dominance for their own product with a monopoly is how you support open and free? Okay, what have Apple, Microsoft, MPEGLA, and all of the industries using MPEG standards been doing for the last 30 years? Last I checked Apple has a market cap 50% greater than Google's, more cash, and still rising while Google has gone sideways for a year on the market, despite unit success with Android (which, again, is dependent on H.264).

Moreover, let's not forget that a few days ago, Google WAS working to make sure it would happen. Youtube uses H.264, Android is dependent upon it and can't forgo it, Chrome supported it... Google was the most effective proponent of H.264, next to Apple, 3 days ago. H.264 is critical to Google WHILE it is trying to kill it.

The Great Re-Encoder in the Sky argument again. (I ask: if Google is appointed the de facto Great Re-encoder in the Sky and are paying for all the decoding licenses annually, and they can give away the vast majority of their products for free because of those monopoly advertising dollars... why don't they just buy licenses for all of the "free" browsers and servers and we all pretend that it was "free" rather than doing all that extra work of encoding, decoding, and storage on principle?)

10. Again, both Apple and Microsoft allow support of system codecs in their browsers. You can describe MPEGLA as "ruthless" and try to paint Apple and Microsoft as choice-haters for having made their own choice for their products, but they are not foreclosing an option. Mozilla has chosen to block system codecs, and Google is removing an optional support they already provided.

11. You say that "VP8 is a Google-controlled proprietary standard" is incorrect, but you only point to a supposed answer. The only way I see it as untrue is that it is a standard. It's a specification, sure, not very well written as I understand it. I'm not sure it's much of what I'd call a standard yet. It's certainly Google-controlled: Google allowing applications to be allowed to make upstream changes is a far cry from open participation. Google makes claims to ownership of patents and the license of the specification so it's certainly proprietary.

Again, Google actually treating WebM as a standard, submitting it to the appropriate standards bodies (ISO and ITU as much so, if not more so, than the w3c) for review and open participation would go a long way in satisfying those who have concerns about the format.

Bottom Line:

Your arguments draw attention to the fact that this is an ideological battle between free software advocacy that solely permits strict adherence to a specification because it requires free usage, in spite of the fact that the real world has never reflected strict adherence to said specification, and those who want a practical and commercial internet while always striving to broaden and provide access to everyone to the best parts of it. (It's quite comical how exceptions for Flash (using H.264) for video and mp3 for audio are made based on ubiquity, necessity, best available... and that these formats are rooted in the same proprietary processes of standardization and governance that H.264 is the 30 year result of, which makes it ubiquitous, necessary, and the best available -- am I really supposed to find Ogg or WebM competitive?)

For many of us, it is perfectly acceptable to say that H.264 should not be included in the w3c's HTML5 specification, but it is completely unacceptable to try to prevent H.264 from having a place on the internet as an option. And it's egregious to actually support and provide the de facto standard as an option and then remove it to prop up an alternative that you own and control.

(Edited for grammatical errors and typos, with some additions. I apologize for any errors and would appreciate any being pointed out, thanks.)

Originally posted by damnbigworld:

For those saying that h264 isn't widespread

Those who have said "h264 isn't widespread" have always ended that sentence with "on the web". Your point has already been addressed in the blog post. See point #6

Originally posted by damnbigworld:

What I don't get is, and I could be wrong, of course, but because Google must license h264 for YouTube they've likely payed the full license cap (my understand is its $6.5mil), so they could easily/legally distribute it.

I'm not sure if Youtube's level of usage would also cover redistribution, but there's also the matter of time - it's an annual licence - and perhaps more importantly of control. MPEG-LA's promise of a "free hit" (using haavard's extremely appropriate analogy) expires in 2016, after which what they charge is entirely at their discretion. Looking at the history of GIF, I think it's pretty foolhardy to think we'll only be using such a video format for the next 5 years and then won't need it any more. 5 years does seem a long time in terms of IT technology, but consider that Unisys' patent on GIF was granted in 1985. If h.264 becomes ubiquitous, MPEG-LA will complete unilateral control over a highly profitable monopoly on web video.

"If h.264 becomes ubiquitous, MPEG-LA will complete unilateral control over a highly profitable monopoly on web video."

And by "unilateral," you mean open to anyone to participate and requiring the agreement of all members (remember, not all members are patent holders) and licensors. And by "highly profitable monopoly," you mean everyone who pays a license is doing so voluntarily to reap a greater derivative profit for themselves and to the benefit of all downstream.

It's pretty hard for any one company to go berzerk and vastly alter the cost of licenses affecting every other member of the group: the members have their own intra-group pressures to both maximize profit while maximizing usage; market realities encourage producers to make as much money as they can, but consumers and demand equally pressure them to keep prices low; eventually the patents expires... The reality is licensing costs are only likely to go down (excepting for inflation). The appropriate comparison is .mp3 licensing where there is massive amounts of non-enforcement (enforcement of the long-tail of users who do not produce profit is more costly than revenue from lost license fees), and the costs have progressively declined over the years. The idea that MPEG licenses get jacked in 5 years is a scary bedtime story -- Boo!

As for the burden of cost: Google is a $200 billion dollar company that gives away search, its mobile operating system, its web operating system, hardware for said mobile and web OS to 1000s at periodic, annual events, all of its applications, a video format supposedly competitive with the "expensive" one that the rest of the world chooses to use, encoding into multiple formats (including the "expensive" one in question) and storage of the world's videos, mapping, navigation, translation, the world's books, and many other services for free; acquires 10s of companies annually at inflated premiums no other company will pay; and awards key employees 3-5x competitive salaries to keep staff from leaving for the competition (again, your champion of openness and freedom!). 6.5 million annually is not a burden. They can pay it for themselves, Mozilla, Opera, and all the independent, profit-seeking individuals who wish to serve content at a far lesser rate than the cost burden they would like to place on consumers for their own benefit.

(Edited for grammatical errors and typos. I apologize for any errors and would appreciate any being pointed out.)

Putting aside for a moment the semantically empty argument over relative "openness", it's kind of presumptuous to call WebM a "standard." Equally laughable is the notion that H.264 will even matter after 2016. But the best joke of all is associating MPEG-LA, a consortium, with words like "monopoly" and "unilateral" when in reality it is Google that is acting in a unilateral fashion.

Being stuck with draft-1 VP8 for WebM and WebP, for 5-10 years, would be like getting stuck with IE6 dominance again. More bandwidth, less quality, wholly unsuitable for certain work because there's only one encoder with few options.

Google needs to build a standard for the bitstream, and other major components, then fix VP8 until it HITS those benchmarks. Then other teams will hit those same benchmarks with their OWN encoders.

h.264 extracts a fee from various forms of distributor, this is certainly true. However the bitstream spec is excellent, and there are many coders which comply with different parts of the standard. So we're stuck: VP8 is kind've crappy and limited, while h.264 makes it difficult for small businesses to compete in media distribution.


Truth be told, the drastically uprated modern Theora could be competitive here. Paired with another Matroska-based container similar to WebM, Theora should perform smoothly and with better size/quality ratios. PsyOpts helped immensely.

I assume Google thinks it can support VP8 with Android 3.0 devices. Now if only Youtube didn't have such a grainy Ogg encoder......

Originally posted by lucideer:

Those who have said "h264 isn't widespread" have always ended that sentence with "on the web". Your point has already been addressed in the blog post. See point #6

No, it hasn't. H264 is widespread on the web.
"6...Sites like YouTube will convert and compress videos anyway"... Yes, to h264, in flv and mp4 containers. Youtube, Facebook, Netflix, Vimeo, Blip, Flickr? All serve h264 video, at what point is something widespread on the web?

Originally posted by lucideer:

consider that Unisys' patent on GIF was granted in 1985

To me that's somewhat erroneous, GIF was a fundamental building block of the web, as we know it. h264 and WebM are not, they are just the current flavor of putting video on the web. As was mpg, mpg2, h263, real, vp6, the many flavors of wmv, were before them, we don't know what's around the corner, hopefully something better than either. There are still times when a properly optimized GIF, is better than a PNG. Generally, you can't say that about compression codecs in video, their lifespan is shorter.

I'd like to thank Tim Faulkner for stating what I've been trying to say for a bit, and in a much clearer fashion I think.

Originally posted by lucideer:

If h.264 becomes ubiquitous, MPEG-LA will complete unilateral control over a highly profitable monopoly on web video.

The fact that MPEG-LA is in charge of h.264 in no way means that there is unilateral control. Unless, that is, the WebM group has unilateral control over vp8. Both are composed of a group of companies. No single company is able to dictate MPEG-LA's actions and many actively compete against each other.

Originally posted by hellspork:

Being stuck with draft-1 VP8 for WebM and WebP, for 5-10 years, would be like getting stuck with IE6 dominance again. More bandwidth, less quality, wholly unsuitable for certain work because there's only one encoder with few options.

Yes, I agree, Google has been rushing vp8 out the door way too fast. Which is odd to say considering how long vp8 had been in development at On2 but then On2 did not do a very good job with vp8. The spec had bugs in it(!) when it was frozen and released. That's because the the spec was basically defined as the encoder/decoder software as they existed when frozen and that software had bugs in it, besides making the spec vague for 3rd party developers to implement. The corrections to the spec exist in an experimental branch for a future vp8.1 or vp9 but then that media will have to replace all of the vp8 media! By then, h.264 will start to fade and be replaced by h.265 which aims to use half the bitrate of h.264. During this whole period the internet will have to settle for a less efficient codec (higher bandwidth costs for everyone) based on a buggy spec that was rushed out (the bugs aren't hard to fix), similar to how people had to settle for an internet experience dictated by the faults in IE6.

I have to admit that I actually miss Ogg Theora in this article. If WebM is discussed as an open alternative, than we should not forget about this more mature and high quality codec.

Great move from Google. h264 should not be online. Also, mp3 and gif must go away. Because ogg and apng is open, free, and have better quality. We must not allow again, from the start, another mistake of using shareware codec.

Originally posted by SteveKong:

I have to admit that I actually miss Ogg Theora in this article. If WebM is discussed as an open alternative, than we should not forget about this more mature and high quality codec.

Maybe you have some proofs that theora is higher quality than webm(vp8)?

Originally posted by damnbigworld:

I'm the Web Director for my company, so making the process more streamlined is very important to me.

So, that looks like your problem?

Originally posted by Asires:

Maybe you have some proofs that theora is higher quality than webm?

http://x264dev.multimedia.cx/archives/541
vp8: http://x264.nl/developers/Dark_Shikari/imagecoding/vp8.png
theora: http://x264.nl/developers/Dark_Shikari/imagecoding/theora.png
Look at the grass o.O That's the difference between having psychovisual optimizations which theora started to add, and optimizing for PSNR like vp8 currently does.

Originally posted by damnbigworld:

For those saying that h264 isn't widespread. Please be aware that it is
-The video format of choice for most large video sites, including YouTube, Netflix and Vimeo
-The defacto format for video podcasts
-Part of Blu-Ray (granted that format years are numbered)
-Supported by billions of devices

They convert whatever video you upload tot them. Why would it matter that online it's webm instead of h264? It's still the same video... And the Blu-Ray and the billions of devices have nothing to do with browsers. Note that here we are talking about browsers. Chrome is a browser. Google is (finally) dumping h264 codec from their browser. They didn't say anything about android, etc...

Originally posted by TimFaulkner:

Why don't they just buy licenses for all of the "free" browsers and servers and we all pretend that it was "free" rather than doing all that extra work of encoding, decoding, and storage on principle?

* That still wouldn't make it open and free.
* Because it's not only the browsers that will have to pay, but soon also basically everyone trying to convert/record a video in h264. You can see it as a time trial. Now you can use it for free, after a certain set of days (starting 2015 or 2016 or something), you'll start having to pay...
* Also (one of) the points is that the online stuff should exist out of 'free' stuff. since h264 isn't really completly free, it shouldn't be online...at all.

Originally posted by TimFaulkner:

try to paint Apple and Microsoft as choice-haters for having made their own choice for their products, but they are not foreclosing an option.

So only supporting h.264 (by default) is giving the users a lot of choices?

Isn't it funny that the only browsers that still support H264 are those made by Companies that have financial stake in the whole H264 thing? I certainly think so.

I admit that the move Google just made isn't the most comfortable for us users or web companies, BUT in the long run, it is certainly the best. Asires makes the best point here, imo.

Originally posted by Asires:

Great move from Google. h264 should not be online. Also, mp3 and gif must go away. Because ogg and apng is open, free, and have better quality. We must not allow again, from the start, another mistake of using shareware codec.

I disagree not with your conclusion, but with your reasoning.

You seem to have started your argument with nothing more than the conclusion that H.264 is "closed", and trivially drew your conclusion from there.

For example, I really have no idea how you can argue against point #10. Even on this one point, can you explain to me how Google offering H.264 would *limit* users' choices? H.264 is a choice itself.

Originally posted by twilightomni:

For example, I really have no idea how you can argue against point #10. Even on this one point, can you explain to me how Google offering H.264 would *limit* users' choices? H.264 is a choice itself.

Well, by helping H.264 become the biggest and therefore standard codec, it would limit the freedom one has to exploit the greatness of the codec because it would cost royalties.

Originally posted by TimFaulkner:

WebM is not an open standard

Originally posted by amigosito:

it's kind of presumptuous to call WebM a "standard."

I don't know why the supporters of a closed web keep pointing out that WebM is not a standard. The actual blog post clearly states that it isn't. So why repeat something no one denied?

Originally posted by twilightomni:

You seem to have started your argument with nothing more than the conclusion that H.264 is "closed", and trivially drew your conclusion from there.

But that's what the whole discussion boils down to. The Ars Technica article claims to talk about "openness", but starts out with a bunch of other red herrings.

The question of openness is addressed clearly, as it should. H264 is not open. It is in direct violation of the W3C patent policy.

The other points are just irrelevant red herrings that were addressed anyway for some reason.

What do you mean by open?

I don't care for the term. Do you mean open-source? There are open-source encoders and decoders for H.264.

What you really mean is royalty-free. What you really mean is violation of the W3C patent policy.

Those are not arguments for open source. They are arguments for a royalty-free content platform for the web.

I have nothing against a royalty-free web. But I also want the right to use the products and services of businesses that do not support a royalty free web. It is Google that is in offense here.

Because in addition to pushing its own noble royalty-free video initiative, it feels the need to also torpedo a video standard that actual content businesses would support.

Originally posted by ernesteban:

Well, by helping H.264 become the biggest and therefore standard codec, it would limit the freedom one has to exploit the greatness of the codec because it would cost royalties.

...Because H.264 costs royalties, Google's support of H.264 to become a standard would limit your freedom to exploit H.264?

That literally doesn't make sense.

Originally posted by ernesteban:

Isn't it funny that the only browsers that still support H264 are those made by Companies that have financial stake in the whole H264 thing? I certainly think so.

Isn't it funny that the only browsers that support WebM are those made by Communities that have an ideological stake in the whole WebM thing?

No, it's not. It's perfectly rational.

Originally posted by hellspork:

Being stuck with draft-1 VP8 for WebM and WebP, for 5-10 years, would be like getting stuck with IE6 dominance again. More bandwidth, less quality, wholly unsuitable for certain work because there's only one encoder with few options.

This is directly misleading, and sounds a bit like... FUD.

The encoder is being improved. Alternative encoders can be created.

Quality is not really an issue.

Bandwidth won't be noticeably worse either (depends on how videos are encoded).

All of this to support a closed video format? I don't like it.

Originally posted by twilightomni:

...Because H.264 costs royalties, Google's support of H.264 to become a standard would limit your freedom to exploit H.264?

That literally doesn't make sense.

No, the royalties would limit my freedom, and the only way of avoiding that is by avoiding the codec altogether (because MPEG-LA will never make it royalty free), and that's what Google is doing.

I'm not sure how that is hard to follow.

Originally posted by twilightomni:

Isn't it funny that the only browsers that support WebM are those made by Communities that have an ideological stake in the whole WebM thing?

No, it's not. It's perfectly rational.

Yeah, but they don't impede those that don't agree with their ideologies. MPEG-LA impedes everyone that don't agree to pay them.

That is what I don't like about using H.264 as the HTML5 video standard.

Originally posted by TimFaulkner:

And by "unilateral," you mean open to anyone to participate and requiring the agreement of all members

By "unilateral" I mean each and every licencee of h.264 will be answerable to the MPEG-LA patent-pool. Patents are a legal matter, anyone leveraging a patented technology are answerable to the patent holder regardless of the agreements of non-patent-holding MPEG-LA members.

Originally posted by TimFaulkner:

And by "highly profitable monopoly," you mean everyone who pays a license is doing so voluntarily to reap a greater derivative profit for themselves and to the benefit of all downstream.

You've contradicted yourself in a single sentence. "doing so voluntarily to reap a greater derivative profit" - if it's necessary to profit it's not voluntary, it's necessary, as a duty to shareholders.

Originally posted by TimFaulkner:

the members have their own intra-group pressures to both maximize profit while maximizing usage

Exactly. In a monopoly situation, maximising usage is no longer necessary (as it is inherent to the system) and is deprioritised in favour of maximising profit.

Originally posted by amigosito:

Putting aside for a moment the semantically empty argument over relative "openness"

Firstly, I love how you immediately "put aside" the primary argument. "Semantically empty"? Why? Disregarding the main reason to use WebM because it doesn't suit you to attempt to rebut it isn't a particularly strong way to get your point across.

Originally posted by amigosito:

it's kind of presumptuous to call WebM a "standard."

Noone is calling WebM a standard. Try rereading the blog post - see point #2

Originally posted by amigosito:

words like "monopoly" and "unilateral" when in reality it is Google that is acting in a unilateral fashion.

How so? They're promoting a format that it will never cost anyone to use. That everyone is absolutely guaranteed to always have complete freedom to encode and redistribute art in, without any fear of legal redress. A format that Google will never directly* profit from.

*obviously they'll indirectly profit from it in the form of advertising, as they do from all of their activities, but that has nothing to do with the relative openness of the chosen format.

Originally posted by damnbigworld:

To me that's somewhat erroneous, GIF was a fundamental building block of the web, as we know it. h264 and WebM are not

Why? Your statement is empty without justification? What makes GIF fundamental and not video?

Originally posted by Astrophizz:

The fact that MPEG-LA is in charge of h.264 in no way means that there is unilateral control. Unless, that is, the WebM group has unilateral control over vp8.

Noone in the WebM group has any "control" over WebM as it is un-patented. That's the differentiation.

9. The market share of browsers that support H.264 exceeds WebM capable browsers"

Google's online advertising monopoly is working on overdrive to ensure that won't happen. If I am not mistaken, the share of open standards based browsers is growing at the expense of Internet Explorer. Although browser market share is impossible to measure reliably, most of the data seems to confirm that.

Actually that statement is completely incorrect. The market share of browsers that support WebM is larger than those that support H.264.

http://gs.statcounter.com/#browser-ww-daily-20101201-20101201-bar

According to Statcounter - in December 2010 - the global market share for all browsers is as followed:

IE - 46.94%
Firefox - 30.76%
Chrome - 14.85%
Safari - 4.79%
Opera - 2.07%
Others - 0.58%

Let’s split those up into h.264 camp and WebM camp:

Browsers that can run WebM videos right now - (Chrome and Opera) - 16.92%
(When Firefox 4 is released this will potentially go up to - 47.68%)

Browsers that can run h.264 videos right now - (Safari) - 2.07%
(When IE9 is released this will potentially go up to - 51.73%)

However the above numbers does not take in account browser upgrades - for example if a user is on Opera 9 then it doesn’t do video tags.

So let’s look at browser upgrades here:

http://royal.pingdom.com/2010/11/05/modern-web-browser-adoption-better-than-expected-71-run-latest-version/

As that show, Chrome is at top with 90.6% on the latest version, followed by Firefox with 80.5% and Opera at 77.2%.

The top 3 browsers that are the most likely to be upgraded to the latest version are all WebM-capable browsers (except for Firefox4 which isn’t released but with 80.5% upgrading, it would not take long for the bulk of Firefox installation to be on latest) and are the most likely to be on the latest version that support WebM video

Safari is at 71.1% and IE is at a low 60.2% so there won’t be as rapid uptake of IE9 when it come out. IE6 is STILL in use for one (I know I know…company policy and all that!)

The upgrade for IE9 would be much lower since it’s not supported for Windows XP (which does command quite a large market share still, bigger than Windows Vista and 7)

So it would suggest that in the near future, the majority of browers that support the video tag will be a WebM majority, not h.264 and the market share of ALL browsers that support H.264 will not even be anywhere near even 40%.

Remember this is about browsers. Embedded hardware etc doesn’t matter here - they don’t dictate the web video market - they don’t do a thing when browsers can’t play h.264.

Right now, the browsers that can play video in WebM format outnumber the browsers that can play video in h.264.

Furthermore - IE9 can support WebM in a video tag BUT the user must install the codec themselves before this happens (MS won’t ship this themselves) - so if IE9 users head to YouTube, Google can automatically provide them with the WebM codec as a installable codec and then IE9 will support WebM so it would actually push the use of WebM codec even more and Safari would actually be the one standing alone here with no WebM support.

I’m aware of Microsoft providing a h264 plugin for Firefox - but it’s more likely that people going to YouTube will get the WebM codec pushed onto them than the h264 plugin getting installed.

But as stated above - right now, browser support for WebM video vastly outnumber support for H.264. And this gap will just increase. WebM is already ahead of H.264.

And people claim that doing this will prevent innovation - I would like to remind you all of a small simple product that have done much to shake up the browsers and give us the Internet today - Firefox.

Firefox will not support h264 - it is a free product, with open source code available to all. They will not pay for h264 licences. Innovation cannot happen if you have to pay vast amount of money for it.

This is why WebM is a good thing. People can innovate new ideas, products and not worry about the cost.

Film makers can create movies and MAKE money from it and not worry about licensing costs from h264 - thus allowing any home-brew projects to experiment and hope for success without worries of cost.

(In case you’re not aware, if you purchase a digital camcorder or camera that does video, for example I own a Canon 500D which can do HD video. The licence for the camera state clearly that the h.264 codec is for personal use only, I cannot stream it etc. This applies even to the high end professional HD camera including RED, Sony, JVC, Panasonic etc. You MUST purchase a licence for H.264 if you wish to stream the movie (on a page with ads to make money for example) or sell it. - that stifle innovation, not encourage it.)

Hardware support for WebM? It’ll come. There are already hardware starting to appear (PowerVR for one) and graphics card (AMD and NVIDIA and Intel for example) will add support soon enough.

>royalties
up to 10 cents per device lol
Guys, this is the Apple-Google battle, not our own.

Originally posted by Asires:

So, that looks like your problem?

Absolutely, which is why when a major company does something that broadly effects the industry's workflow detrimentally, at least in the short term (1-2 years), it should cause concern. Which it hasn't among the rah-rah WebM supporters. I support WebM and Googles push to utilize it, but they should have done it through YouTube rather than dropping h264 browser support. WebM in its current state is a very immature technology, h264 is not. As web professionals, we are directly effected by the tools we have at our disposal, the problem being there aren't currently sufficient tools to support it.

Originally posted by lucideer:

Why? Your statement is empty without justification? What makes GIF fundamental and not video?

Point missed! Video is fundamental, a specific video format/codec is not. Codecs don't last on the web, I pointed out several examples that have been supplanted by h264. There are essentially 3 raster image filetypes for the web gif,jpg, and png. They are essential to the web, the first two more than the last since since IE PNG support is still not great through v8. Video is essential, as well, the difference is that the format of the video does not stay constant for long. Flash-based video has won the race for a long time, but that has went though multiple codecs since the beginning to finally arrive at the industry standard of h264 in various containers. We hope to eventually eliminate the need for Flash for Video, WebM appears to do the exact opposite, at least in the short term.

Originally posted by amigosito:

it is Google that is acting in a unilateral fashion.

On the contrary. They have companies and organizations like Opera and Mozilla backing them, and basically anyone who wants an open web instead of a patent-encumbered web.