The Ambiguity of Open and H.264 vs. VP8
antimatter15.com
antimatter15.com
>VP8 is not a standard. It was developed secretly by a single company, and until recently, had only a single working implementation. The public wasn’t open to collaboration on the specification until the bitstream spec was frozen, including the bugs that existed within.
This is an interesting point. One I had never even realized.
> Its severely bound in what it can be by trying to avoid
> H.264's existing patents.
Yeah probably - ask the Theora guys how hard its been ;-) The upside is that it could very much, out of necessity, drive some considerable advancement in video compression algorithms.
Since H264 was implemented in the open with open and large membership to the body, it is more difficult for some company to come out of the woodwork and say, "Hey, we have a patent that H264 rides on, but we didn't know it until now" Note, QCOM tried this with H264 several years ago and got their butt handed to them by the courts (http://www.broadcom.com/press/release.php?id=1037466). People have had 10 years to examine this open specification. I think its reasonable to say that one would have found a violation by now. And almost all the majors are already either licensees or licensors. There's a slim chance that there's a patent floating around in some patent troll portfolio that hasn't been examined, but that's the only real vector I can see of a legitimate surprise H264 patent violation.
Whereas with WebM its been in the public domain less than a year. It could feasibly take a few years for it to be reaonable that anyone who would have done due diligence to have done so. It's completely reasonable that in a year from now Siemans comes out and says, "WebM infringes these three patents we have". A court is not likely to throw it out saying they've been sitting on them.
This is the true genious of the WebM license. It means even if this does happen in the future, Google can destroy whoever has this patent through a massive counter. It also means that any challenge would have to be made extremely carefully.
edited for spelling :)
One area that unfortunately it didn't really cover was the impact of widely available hardware and GPU-accelerated decoding for H.264. Surely the millions of non-PC devices being sold with H.264 support (even baseline) will have some effect on the outcome of the new video format war.
the infrastructure argument is much stronger.
Would a person believe Microsoft's good intentions if it was WMV instead of WebM?
This is the key point that everyone's missing.
>Google hereby grants to you a perpetual, worldwide, non-exclusive, no-charge, royalty-free, irrevocable (except as stated in this section) patent license to make, have made, use, offer to sell, sell, import, transfer, and otherwise run, modify and propagate the contents of this implementation of VP8, where such license applies only to those patent claims, both currently owned by Google and acquired in the future, licensable by Google that are necessarily infringed by this implementation of VP8. This grant does not include claims that would be infringed only as a consequence of further modification of this implementation. If you or your agent or exclusive licensee institute or order or agree to the institution of patent litigation against any entity (including a cross-claim or counterclaim in a lawsuit) alleging that this implementation of VP8 or any code incorporated within this implementation of VP8 constitutes direct or contributory patent infringement, or inducement of patent infringement, then any patent rights granted to you under this License for this implementation of VP8 shall terminate as of the date such litigation is filed.
It explicitly grants you the ability to modify the library in whatever way you want. This naturally includes a complete re-implementation.
There's a strangely worded section in there, but it is a clear cause-and-effect:
>This grant does not include claims that would be infringed only as a consequence of further modification of this implementation.
So, any claims infringed only as a consequence of further modification are not covered. But if the patent claims were already infringed by Google's implementation, then you are covered.
Q: What if someone makes a change to the code and gives it to me. Do I have a patent license from Google for that change?
A: You still have the right to redistribute but no patent license for the changes (if there are any patents covering it). We can't give patent licenses for changes people make after we distribute the code, as we have no way to predict what those changes will be. Other common licenses take the same approach, including the Apache license.
I'm not sure if it supports your statement, it would be great if someone more knowledgable about this can clarify.
Cameras shoot in HD and realtime compression isn't very efficient, so such raw files are unsuitable for the web.
For the web you still need to lower resolution and bitrate of the files. Re-encoding step is necessary, and it allows format of web video to be independent of source format.
o They run the risk of patent infringement
o H.264 is already playable everywhere, universally,
either natively (and with hardware support) or with Flash.Same goes for all the other manufacturers- they are already MPEG-LA licensees and safe from the vast majority of patent threats.
So all those implementing WebM still needs to license H.264 to be safe from patent threat, and all those implementing WebM almost all have implemented H.264. WebM is only an addition to H.264. So the question for Google is always why dropping H.264 and never why adding WebM (since it is always there).
http://fosspatents.blogspot.com/2010/06/mpeg-las-avch264-lic...
hence the <video> situation we were in even before last week and google's blog post.
They are after all a government sanctioned cartel. You don't get away with this kind of anti-competitive price discrimination otherwise.
Which is why it always confuses me that people talk about what an excellent codec it is. How do you know it's excellent when the legal structures are set up specifically to exclude competition, and therefore comparison?
Great article though!
It could be and its been challenged at least once in the past (however Qualcomm lost that challenge), but though its not free, its patent restrictions are pretty clear and tested. It is 7 years old. While stranger things have happened, its very unlikely that as a licensee of H.264 you are going to be sued.
WebM's status is not as clear, sometimes you choose the devil you know.
Best YouTube user experience ever, auto loads the highest quality every time, and downloads faster. This is the way YouTube should be!
I was under the impression that they all used Flash and Silverlight because then they could implement DRM. They don't need to do that on iOS devices, because the whole platform is DRM'd.
For example BBC iPlayer is HTML5 to iOS, but Flash to Mac OS X. They've announced that all the other mobile devices will be served via Flash.
People that want to post videos of their cat dancing or a backyard physics experiment on the other hand, have totally different needs, like the ability to post it in their subscriber only cat physics forum without worrying about licence fees.
(I'm not being rhetorical)
It's too simplistic to argue H.264 as being the best for the user, hence it should be supported. Figuring out whether it is actually best for the user depends on a lot of variables, a lot of hard work, and some measure of work.
Back in the day when IE dominated Netscape Navigator/Communicator, IE was the best end-user experience. Flash-forward a few years and IE is languishing, with practically no new development happening on it. At this time (prior to Mozilla taking off), IE is still the best end-user experience available, but not the best end-user experience possible. It to took the Mozilla Suite and it's spin-off, Firefox, to bring a better user-experience to market.
The same applies now. While h.264 might provide the best experience right now, it's (possibly) at the cost of the experience in the future. What happens in the future when the MPEG-LA decides to start changing/enforcing royalties? What growing pains will the industry have to endure to change formats away from h.264 if that becomes an issue? What if h.264 is so entrenched that they can't migrate away from it? Why not just standardize on something now that won't have those problems later?
To fall back on the analogy, if IE hadn't pushed Netscape out of the market, what would the web ecosystem look like now?
However, there is one rather large oversight:H.264 is an open standard
This is not necessarily correct. As a citizen of the EU, I have to disagree. One of the EU pre-requisites[1] for something being qualified as an Open Standard is: The intellectual property - i.e. patents possibly present - of (parts of) the standard is made irrevocably available on a royalty-free basis.
[1]: http://en.wikipedia.org/wiki/Open_standard#European_Union_de...
That's not true, unfortunately.
There may have been no lawsuits (I can't comment on that, though I'd like to see your data). But that's just because the MPEG-LA hasn't felt like suing people. If you're shipping an H.264 encoder or decoder in binary form to people and not paying the relevant licensing fees, then you're infringing on the patents. There's no exception for "non-commercial" use there last I checked. If you think there is, please cite.
Somebody will come up something similar to the hack/exploit that Mozilla did: set up a for-profit corporation to receive money and a non-profit foundation (which owns 100% of corp's shares) to do the development. Then they can claim that the non-profit foundation is the one that's really using the patents and hence should not have to pay royalties.
Google's, Mozilla's and Opera's actions and careless, uninformed (or possibly intentionally slanderous) responses in this discussion clearly show that their issue with H.264 isn't that H.264 isn't open enough, but rather that it isn't gratis enough.
>incompatible with open source
Does this make sense to anyone? It seems to me the word open doesn't mean much now.
I've never found the argument you're making very compelling. It's "open" in the same way that golf course with a $500 greens fee is open.
What it really constitutes is a way for existing market players to collude, and thereby exclude future participants. Some of those future participants might be quite disruptive, and benefit consumers.
It's really pretty funny that some terribly corrupt standards organizations arrived at a totally unsatisfactory definition of "open", and then tried to claim that better definitons of "open" somehow infringed(!) on their early, bad definition.
Now I know your point wasn't actually about golf, but a lot of "open" facilities do cost money to use. It's just that they're non-discrimanatory about their use and charge what they consider reasonable fees (enough to pay for staff, upkeep, improvements, etc...).
This is actually very similar to MPEG-LA, the big difference being that MPEG-LA is not non-profit. But then again neither is Google. What would be a nice move by Google is to actually give all On2 property to Mozilla or Xiph. Or ever better yet, Outercurve (to show that they don't really want to indirectly exert disproportionate influence).
I don't think you read what I wrote, and I doubt you play golf. The fee at a decent public course where I grew up in Oregon is $37.
> This is actually very similar to MPEG-LA,
It's not at all similar. MPEG-LA collects rent on imaginary property.
So MPEG in this light can be seen as a collusive monopolistic oligarchy, where if you're in the club you get something for free that everyone else has to pay for. Not very competitive.
A bit of history: early on, On2 made a stink about the monopolistic nature of MPEG-LA:
And anyone can go to each patent holder and negotiate licensing of patents since the licensors still hold the patents. Part of the reason that MPEG-LA exists is to make it so that you don't have to negotiate with fifty companies in order to implement H264. MPEG-LA doesn't block you from doing this if you prefer that route, but it will take a LOT more time and likely be more expensive.