8 ms·

I have to disagree, the article points out a lot of flaws in the Ars. article. The premise of the article is self-contradictory: H.264 is patent encumbered, ope

I have to disagree, the article points out a lot of flaws in the Ars. article. The premise of the article is self-contradictory: H.264 is patent encumbered, open web means open/free to use for anybody, these two don't relate to each other. I find it dubious to suggest the following: we should use it, just because most other people use it. Google dropping this codec actually helps prevent h.264's monopoly.

redial 16y ago
"Google dropping this codec actually helps prevent h.264's monopoly." How about a Google monopoly? At least h.264 is owned by a number of different companies. Never mind that almost everyone agrees that it is the better codec.
- Pewpewarrows 16y ago
  You can't have a monopoly on a product like WebM that's now open sourced.
  - recoiledsnake 16y ago
    In real life you easily can, because of incompatibilities with the existing software. Eg. Chrome and Android are technically open-source but their direction and development are completely governed by Google. If you make changes and are unable to convince Google to incorporate them, the video generated by your software will be incompatible with all the millions of WebM decoders out there.
    - zppx 16y ago
      I do not care about WebM or H.264 since I dislike both, my comment will concern the license and the situation. WebM have a Apache style license, you can grab the source, fork it and develop your codec, and you're not even obligated to open a repository including your changes, they can remain closed. Google will not stop you from doing so and guarantees that he will not sue you asking for royalties, since a license like the Apache license gives the use of the patent for free. The case of Android and Chrome is simple, Google employes the majority of developers in both projects, however Google does not employes the majority of Linux kernel developers an as a result much of the Android code isn't in the kernel.org code. Similarly the directions of Webkit is decided by the two main committers: Google and Apple.
      - recoiledsnake 16y ago
        I am talking about practicality or the chances of it happening. Anyone can do what they want, but if it's incompatible with the 800lb gorilla's offering, it's basically a moot point.
      - Locke1689 16y ago
        Actually, recoiledsnake's scenario has already happened. The VP8 "spec" isn't a spec at all, it's just copy pasted C code. Unfortunately, they also chose to reuse the same code for the encoder and the decoder. This means bugs in either one can propagate into both. There was already one of these bugs found, fixed, and had a patch applied to the source tree. Google came in and reverted that patch because too much video had already been encoded with the bug in the encoder. This is exactly what happened with IE6, where bugs in the product become de facto requirements on the web. Maybe the H264 spec drones on and on but at least it's a spec. The VP8 "spec" is just a mandate from Google. A shitty one, at that. I'll take the small risk of patent litigation among only the biggest players for an open specification and standard any day. Of course, there's also the fact that H264 is way better than VP8 which it seems like no one cares about.
        true_religion 16y ago
        It was a 'bug fix' that caused a major regression. What would you have them do?
        Locke1689 16y ago
        Write a spec, not a "bitstream guide," so that this problem doesn't happen in the first place? Or at least admit that your spec is not finalized and is, in reality, a piece of crap that needs wide peer review first.
  - redial 16y ago
    Like GCC or WebKit or Java? Being open source does not mean nobody owns it.
    - jarek 16y ago
      Who has a monopoly on Webkit?
- spiffworks 16y ago
  What does owning a video codec even mean if you make it royalty-free? Is there a concern that if webm becomes widely used Google will use its control of the code to harm competitors? In that case, you should know that the webm project is in fact run as a separate project from Google[1]. [1]http://www.webmproject.org/about/supporters/ http://www.webmproject.org/about/supporters/
- haberman 16y ago
  > How about a Google monopoly? At least h.264 is owned by a number of different companies. It's amazing to me. In the 90s Microsoft's overt policy was "embrace, extend, extinguish," and it called the GPL a "cancer," the MPAA was launching war against DeCSS, and Amazon.com was flexing its one-click patent against competitors. Open source developers had no powerful allies. Now you have a powerful company like Google that releases millions of lines of open source, sponsors college students to work on open source, fights for openness standards in wireless spectrum auctions, fights for open standards like HTML5, open sources its entire mobile operating system, and in its latest move, buys a video codec company for $106 million and promptly pledges all of the patents for use by anybody. And yet some people would prefer the codec whose rights are effectively owned by a known patent troll (MPEG-LA) because "at least [it] is owned by a number of different companies." It goes to show that you can never please everyone.
  - Locke1689 16y ago
    I don't know what kind of fantasy world you live in where the only reason you could like H264 is because it's patented. I like H264 because VP8 is an inferior technology.
  - recoiledsnake 16y ago
    Why doesn't Google release the source code of the flavor of Linux that it uses after earning tens of billions basically leveraging Linux code? Google open sources things that are not central to it's business, just like every other company out there. MPEG-LA cannot be a patent troll let alone a known one, because it's a loose organization of companies that agree on a standard to prevent fragmentation of video and let people deal with one entity instead of a thousand. Those individual companies could be trolls. >fights for open standards like HTML5 Really? Then this move to remove support for H.264 will get Flash even more entrenched because it would be the easiest way for content providers to get WebM video working on IE9 and Safari. Flash has already announced upcoming support for WebM.
    - irons 16y ago
      MPEG-LA cannot be a patent troll let alone a known one, because it's a loose organization of companies that agree on a standard to prevent fragmentation of video and let people deal with one entity instead of a thousand. Also, the licensing terms are fixed at a maximum 10% rise per five-year renewal. Given the lifespan of patents, that's a low ceiling.
eddanger 16y ago
Why not drop Flash too?
- Pewpewarrows 16y ago
  You clearly haven't read the article. Flash is a sandboxed plugin running alongside the browser. It's not the internals of a new and heatedly debated HTML tag, which ultimately has to be an open standard and format, and supported until HTML6 comes around in another decade. Choosing to drop experimental support for a closed, proprietary, and royalty-encumbered codec in a tag's specification that hasn't even been finalized by the W3C yet isn't even in the same universe as dropping support for an external plugin.
  - podperson 16y ago
    How is supporting arbitrary codecs via the <video> tag any different from supporting arbitrary fonts through CSS font-family?
    - nitrogen 16y ago
      Supporting arbitrary codecs is more like supporting arbitrary font formats. Each separate codec uses its own (potentially vastly) different way of encoding video, and requires its own code to handle. Using the operating system facilities is too risky from a security standpoint (such as if someone finds an exploitable bug in the ancient Indeo codec).
- memetichazard 16y ago
  Can't drop something you didn't have in the first place. As many have argued, it's disingenuous to treat Flash and the H.264 as being equal in this argument. One of these would be implemented into the browser rather than a plugin. One of these would incur a royalty cost. One of these has viable alternatives.
  - recoiledsnake 16y ago
    From http://techcrunch.com/2010/06/25/google-chrome-flash/ http://techcrunch.com/2010/06/25/google-chrome-flash/ Google Chrome Now Comes With Flash Built In 391 View Comments Robin Wauters Jun 25, 2010 Last March, Adobe and Google jointly announced that Flash Player would soon come built in to the latter’s Chrome browser, eliminating the need for users to download, install and update it separately.
tomlin 16y ago
Those who seem to be opposed to Google's decision are often in the Apple camp (disclosure: I am in the Apple camp) but are also the same people who wish Flash to die because of it's closed/proprietary nature. So why can't we just envision ourselves embracing H.264 and picture what it'll be like in 10 years? Which camp will be trying to push it out the door because of licensing issues? There will be licensing and royalty issues. MPEG LA is a business at heart. Google is as well, certainly. But WebM could be ours instead of theirs, no? I think what people miss about WebM is that it isn't perfect - and no one is saying it is. It's not fully developed, but shouldn't we participate in fully developing WebM under an open source/royalty-free vision? Can't we all just get along?
- pohl 16y ago
  So why can't we just envision ourselves embracing H.264 and picture what it'll be like in 10 years? You mean imagine when we're 2 or 3 years from the last of the H.264 patents expiring? At that time I would expect it to only be ditched for something that offers substantially better quality/bitrate.
- kenjackson 16y ago
  There will be licensing and royalty issues. When a big company in Redmond says stuff like this, its called FUD. This is FUD, plain and simple. What technologies have patents, fees and/or licensing around them? H264, MP3, Flash, USB, Java, HDMI, CDMA, GSM, LTE, WiMax, Firewire, CD, DVD, BluRay, MP3, AMOLED, and many more. To argue that computing is going to collapse with the existence of all these is pure FUD. Its FUD when MS does it. It's FUD when Google or the Google camp does it. UPDATE: And I forgot a biggie, MPEG-2.
  - tomlin 16y ago
    Not so fast. These licensed technologies you speak of are not embedded within web browser standards. It's a big difference. Flash, for one, is a plugin. It hitches a ride with the browser, it's not part of its standardized, internal workings. Google probably will benefit financially from this change and the rhetoric from Google's camp probably is FUD. No different than what Apple is doing with it's proprietary App Store process or it's closed source OS or how Apple's camp (again, I'm apart of this camp) manages to stick-shift through an uncomfortable conversation about open source beyond WebKit. The difference is that WebM might also help open source as well. Just because it benefits a corporation doesn't necessarily mean it's the devil.
alanh 16y ago
> Google dropping this codec actually helps prevent h.264's monopoly. A baseless assertion. It’s far more likely that most sites will continue using H.264, wrapping it in Flash for desktop users, and then sending it naked to mobile devices including iOS and most other video-capable phones. Economically, what is the incentive to do otherwise, as a publisher? How in the world do you figure otherwise? And can you please fucking stop pretending your own definition of “open” is the same as everyone else’s? Open ≠ free. Sometimes people use it that way. But there IS open source software (GPL’d, even) that is NOT free of cost (like many Wordpress themes & plugins), and there IS open source software that has been tripped up by PATENT TROLLS, and when we are talking about open SPECIFICATIONS the word “open” should be demonstrating that more than one party controls it — which IS the case with H.264 but NOT the case with VP8. So if you want to argue about openness, DEFINE your terms and ACCEPT others will not agree, but we CAN and SHOULD be dealing in terms of facts regarding who controls the spec and who can use it for free, not to mention what’s best for users (something Chrome is otherwise fantastically and fanatically obsessed with).
- chc 16y ago
  I'm pretty sure when most people hear "open specification," they'll think it means "open for anyone to read and implement" or something along those lines, not "owned by a consortium of corporations rather than a single corporation." I know that's what people mean when they say H.264 is open, but "owned by a consortium" is not the usual connotation of "open."
  - mikeryan 16y ago
    I'm pretty sure when most people hear "open specification," they'll think it means "open for anyone to read and implement The shouldn't, thats not what it means. An open specification means that anyone can contribute to the authoring of the specification it has no bearing on licensing terms. Which is the OP's point "open" means a lot of different things.

redial 16y ago

"Google dropping this codec actually helps prevent h.264's monopoly." How about a Google monopoly? At least h.264 is owned by a number of different companies. Never mind that almost everyone agrees that it is the better codec.

Pewpewarrows 16y ago
You can't have a monopoly on a product like WebM that's now open sourced.
- recoiledsnake 16y ago
  In real life you easily can, because of incompatibilities with the existing software. Eg. Chrome and Android are technically open-source but their direction and development are completely governed by Google. If you make changes and are unable to convince Google to incorporate them, the video generated by your software will be incompatible with all the millions of WebM decoders out there.
  - zppx 16y ago
    I do not care about WebM or H.264 since I dislike both, my comment will concern the license and the situation. WebM have a Apache style license, you can grab the source, fork it and develop your codec, and you're not even obligated to open a repository including your changes, they can remain closed. Google will not stop you from doing so and guarantees that he will not sue you asking for royalties, since a license like the Apache license gives the use of the patent for free. The case of Android and Chrome is simple, Google employes the majority of developers in both projects, however Google does not employes the majority of Linux kernel developers an as a result much of the Android code isn't in the kernel.org code. Similarly the directions of Webkit is decided by the two main committers: Google and Apple.
    - recoiledsnake 16y ago
      I am talking about practicality or the chances of it happening. Anyone can do what they want, but if it's incompatible with the 800lb gorilla's offering, it's basically a moot point.
    - Locke1689 16y ago
      Actually, recoiledsnake's scenario has already happened. The VP8 "spec" isn't a spec at all, it's just copy pasted C code. Unfortunately, they also chose to reuse the same code for the encoder and the decoder. This means bugs in either one can propagate into both. There was already one of these bugs found, fixed, and had a patch applied to the source tree. Google came in and reverted that patch because too much video had already been encoded with the bug in the encoder. This is exactly what happened with IE6, where bugs in the product become de facto requirements on the web. Maybe the H264 spec drones on and on but at least it's a spec. The VP8 "spec" is just a mandate from Google. A shitty one, at that. I'll take the small risk of patent litigation among only the biggest players for an open specification and standard any day. Of course, there's also the fact that H264 is way better than VP8 which it seems like no one cares about.
      - true_religion 16y ago
        It was a 'bug fix' that caused a major regression. What would you have them do?
        Locke1689 16y ago
        Write a spec, not a "bitstream guide," so that this problem doesn't happen in the first place? Or at least admit that your spec is not finalized and is, in reality, a piece of crap that needs wide peer review first.
- redial 16y ago
  Like GCC or WebKit or Java? Being open source does not mean nobody owns it.
  - jarek 16y ago
    Who has a monopoly on Webkit?
spiffworks 16y ago
What does owning a video codec even mean if you make it royalty-free? Is there a concern that if webm becomes widely used Google will use its control of the code to harm competitors? In that case, you should know that the webm project is in fact run as a separate project from Google[1]. [1]http://www.webmproject.org/about/supporters/ http://www.webmproject.org/about/supporters/
haberman 16y ago
> How about a Google monopoly? At least h.264 is owned by a number of different companies. It's amazing to me. In the 90s Microsoft's overt policy was "embrace, extend, extinguish," and it called the GPL a "cancer," the MPAA was launching war against DeCSS, and Amazon.com was flexing its one-click patent against competitors. Open source developers had no powerful allies. Now you have a powerful company like Google that releases millions of lines of open source, sponsors college students to work on open source, fights for openness standards in wireless spectrum auctions, fights for open standards like HTML5, open sources its entire mobile operating system, and in its latest move, buys a video codec company for $106 million and promptly pledges all of the patents for use by anybody. And yet some people would prefer the codec whose rights are effectively owned by a known patent troll (MPEG-LA) because "at least [it] is owned by a number of different companies." It goes to show that you can never please everyone.
- Locke1689 16y ago
  I don't know what kind of fantasy world you live in where the only reason you could like H264 is because it's patented. I like H264 because VP8 is an inferior technology.
- recoiledsnake 16y ago
  Why doesn't Google release the source code of the flavor of Linux that it uses after earning tens of billions basically leveraging Linux code? Google open sources things that are not central to it's business, just like every other company out there. MPEG-LA cannot be a patent troll let alone a known one, because it's a loose organization of companies that agree on a standard to prevent fragmentation of video and let people deal with one entity instead of a thousand. Those individual companies could be trolls. >fights for open standards like HTML5 Really? Then this move to remove support for H.264 will get Flash even more entrenched because it would be the easiest way for content providers to get WebM video working on IE9 and Safari. Flash has already announced upcoming support for WebM.
  - irons 16y ago
    MPEG-LA cannot be a patent troll let alone a known one, because it's a loose organization of companies that agree on a standard to prevent fragmentation of video and let people deal with one entity instead of a thousand. Also, the licensing terms are fixed at a maximum 10% rise per five-year renewal. Given the lifespan of patents, that's a low ceiling.

eddanger 16y ago

Why not drop Flash too?

Pewpewarrows 16y ago
You clearly haven't read the article. Flash is a sandboxed plugin running alongside the browser. It's not the internals of a new and heatedly debated HTML tag, which ultimately has to be an open standard and format, and supported until HTML6 comes around in another decade. Choosing to drop experimental support for a closed, proprietary, and royalty-encumbered codec in a tag's specification that hasn't even been finalized by the W3C yet isn't even in the same universe as dropping support for an external plugin.
- podperson 16y ago
  How is supporting arbitrary codecs via the <video> tag any different from supporting arbitrary fonts through CSS font-family?
  - nitrogen 16y ago
    Supporting arbitrary codecs is more like supporting arbitrary font formats. Each separate codec uses its own (potentially vastly) different way of encoding video, and requires its own code to handle. Using the operating system facilities is too risky from a security standpoint (such as if someone finds an exploitable bug in the ancient Indeo codec).
memetichazard 16y ago
Can't drop something you didn't have in the first place. As many have argued, it's disingenuous to treat Flash and the H.264 as being equal in this argument. One of these would be implemented into the browser rather than a plugin. One of these would incur a royalty cost. One of these has viable alternatives.
- recoiledsnake 16y ago
  From http://techcrunch.com/2010/06/25/google-chrome-flash/ http://techcrunch.com/2010/06/25/google-chrome-flash/ Google Chrome Now Comes With Flash Built In 391 View Comments Robin Wauters Jun 25, 2010 Last March, Adobe and Google jointly announced that Flash Player would soon come built in to the latter’s Chrome browser, eliminating the need for users to download, install and update it separately.

tomlin 16y ago

Those who seem to be opposed to Google's decision are often in the Apple camp (disclosure: I am in the Apple camp) but are also the same people who wish Flash to die because of it's closed/proprietary nature. So why can't we just envision ourselves embracing H.264 and picture what it'll be like in 10 years? Which camp will be trying to push it out the door because of licensing issues? There will be licensing and royalty issues. MPEG LA is a business at heart. Google is as well, certainly. But WebM could be ours instead of theirs, no? I think what people miss about WebM is that it isn't perfect - and no one is saying it is. It's not fully developed, but shouldn't we participate in fully developing WebM under an open source/royalty-free vision? Can't we all just get along?

pohl 16y ago
So why can't we just envision ourselves embracing H.264 and picture what it'll be like in 10 years? You mean imagine when we're 2 or 3 years from the last of the H.264 patents expiring? At that time I would expect it to only be ditched for something that offers substantially better quality/bitrate.
kenjackson 16y ago
There will be licensing and royalty issues. When a big company in Redmond says stuff like this, its called FUD. This is FUD, plain and simple. What technologies have patents, fees and/or licensing around them? H264, MP3, Flash, USB, Java, HDMI, CDMA, GSM, LTE, WiMax, Firewire, CD, DVD, BluRay, MP3, AMOLED, and many more. To argue that computing is going to collapse with the existence of all these is pure FUD. Its FUD when MS does it. It's FUD when Google or the Google camp does it. UPDATE: And I forgot a biggie, MPEG-2.
- tomlin 16y ago
  Not so fast. These licensed technologies you speak of are not embedded within web browser standards. It's a big difference. Flash, for one, is a plugin. It hitches a ride with the browser, it's not part of its standardized, internal workings. Google probably will benefit financially from this change and the rhetoric from Google's camp probably is FUD. No different than what Apple is doing with it's proprietary App Store process or it's closed source OS or how Apple's camp (again, I'm apart of this camp) manages to stick-shift through an uncomfortable conversation about open source beyond WebKit. The difference is that WebM might also help open source as well. Just because it benefits a corporation doesn't necessarily mean it's the devil.

alanh 16y ago

> Google dropping this codec actually helps prevent h.264's monopoly. A baseless assertion. It’s far more likely that most sites will continue using H.264, wrapping it in Flash for desktop users, and then sending it naked to mobile devices including iOS and most other video-capable phones. Economically, what is the incentive to do otherwise, as a publisher? How in the world do you figure otherwise? And can you please fucking stop pretending your own definition of “open” is the same as everyone else’s? Open ≠ free. Sometimes people use it that way. But there IS open source software (GPL’d, even) that is NOT free of cost (like many Wordpress themes & plugins), and there IS open source software that has been tripped up by PATENT TROLLS, and when we are talking about open SPECIFICATIONS the word “open” should be demonstrating that more than one party controls it — which IS the case with H.264 but NOT the case with VP8. So if you want to argue about openness, DEFINE your terms and ACCEPT others will not agree, but we CAN and SHOULD be dealing in terms of facts regarding who controls the spec and who can use it for free, not to mention what’s best for users (something Chrome is otherwise fantastically and fanatically obsessed with).

chc 16y ago
I'm pretty sure when most people hear "open specification," they'll think it means "open for anyone to read and implement" or something along those lines, not "owned by a consortium of corporations rather than a single corporation." I know that's what people mean when they say H.264 is open, but "owned by a consortium" is not the usual connotation of "open."
- mikeryan 16y ago
  I'm pretty sure when most people hear "open specification," they'll think it means "open for anyone to read and implement The shouldn't, thats not what it means. An open specification means that anyone can contribute to the authoring of the specification it has no bearing on licensing terms. Which is the OP's point "open" means a lot of different things.