MPEG LA Says 12 Parties Have Essential WebM Patents 136
suraj.sun tips this report from the H Online:
"The hopes that the VP8 codec at the heart of Google's open source WebM video standard would remain unchallenged in the patent arena are diminishing after the MPEG LA says 12 parties hold patents that its evaluators consider essential to the codec. ... No VP8 patent pool has been formed yet; the MPEG LA says it met with the patent holders in late June and is 'continuing to facilitate that discussion' but the decision to form a pool is up to the patent holders. ... Google responded to the MPEG LA's interview saying it is 'firmly committed to the project and establishing an open codec for HTML5 video' and noting the April launch of the WebM CCL, a community cross-licencing agreement for essential WebM related patents."
One single sentence says it all. (Score:4, Insightful)
The parties involved are as yet unnamed and MPEG LA told patent analyst Florian Mueller that "confidentiality precludes [MPEG LA] from disclosing the identity of the owners".
This smells like more bullshit extortion.
Re: (Score:1)
If software patents continue the way they are going, all patent ownership information should all be made 100% public. Especially in the case of litigation.
Re: (Score:1)
All patent information is 100% public. The difficulty is searching through all existing patents to determine which ones are or might be applicable to your work.
Re: (Score:2)
Re: (Score:2)
This smells like more bullshit extortion.
That's a pretty contradictory conclusion, no? To extort anyone, they'd have to show the patents. As long as they don't show them they can not extort anyone.
Re: (Score:3)
This smells like more bullshit extortion.
That's a pretty contradictory conclusion, no? To extort anyone, they'd have to show the patents. As long as they don't show them they can not extort anyone.
This is like saying "something" on your house I thing is mine... But I won't TELL YOU what it is because then yod hand it over or dig out the bill of sale. It's all about "kicking on the door" .. That's extortion.. Look Gogle, that's am awful nice codex you got there.. Shame if somebody started DMCA'ing your followers, "buddy".
Re: (Score:2)
No, it's a warning shot. Eventually the patent holders, should they decide to proceed, will have to bring a case to trial and specifically name the patents in question. To suggest that they can somehow proceed with a legal case without naming such patents, is patently (pardon the pun), ridiculous.
Re: (Score:2)
Thats not how it works though. Remember all that SCO bullshit? SCO where sending around extortion letters to companies threatening to sue if they didn't pay up and when the companies would say "on what grounds?" SCO would reply "Thats confidential". Many companies paid up anyway, because its cheaper to just pay them "gtfo and leave us alone" protection racket money than get in some protracted mud wrestle with a fuckwit litigation factory.
Microsoft are currently doing a similar thing , getting companies to p
Re: (Score:2)
Microsoft are currently doing a similar thing , getting companies to pay patent licencing fees on linux whilst not revealing exactly what patents are being licenced.
Actually, sabre-rattling over Linux or no, Apple and Microsoft are usually pretty clear which patents Android is violating (they're pretty much listed in every news article talking about it). The fact that many of these patents shouldn't even exist, by the way, is completely irrelevant (I love living in a country where the patent office has a specific directive that computer software may not be patented, with no higher patent office to just ignore them).
Re: (Score:2)
SCO was about copyrights, not patents. Entirely different things.
Re: (Score:2)
No, to sue someone they need to show the patents. To extort, they just need to make threats.
This is the patent equivalent of vague threats like: If you don't pay us, we are going to hurt you... We have ways of hurting you you can't even imagine..
Re: (Score:2)
You actually think any company would pay another company without knowing what they are paying for, and without getting any kind of contrat that specifies what exactly they are getting?
Re: (Score:2)
Exactly what they are getting is easy; pay up and we won't sue you over any patents that may cover "x". It will be cheaper for you than working out a licensing deal in court.
Re: (Score:2)
And the fact that the MPEG-LA has never done anything like that is clearly not any kind of obstacle for you believing this?
Re: (Score:2)
It's an attempt to create FUD to stop people using it. Microsoft did the same thing to linux years back with it's "256 patents in linux" announcements which eventually melted down to 13 patents that they refused to tell anyone which ones exactly.
Re: (Score:2)
Which is an entirely different thing than "extortion".
Re: (Score:2)
At the time Microsoft where trying to sell patent license agreements.
Re: (Score:2)
Where did you get that crazy idea?
Re:One single sentence says it all. (Score:5, Insightful)
patent analyst Florian Mueller
Sorry, I just felt the need to quote that. It made me laugh so hard I almost dropped my laptop.
Re: (Score:2)
Extorting means making a demand. So far, there has been nothing but a press press release about MPEG-LA companies finding patent violations. And the companies have not had their names announced yetso what. That is probably to keep Google from getting to them before then can mount an attack. There is nothing wrong with that.
As to software patents themselves, I'm conflicted. While on one hand, I think that people that come up with original, non-evolutionary ideas should be paid for them. On the other hand, I
Why has this announcement taken so long? (Score:2)
Re:Why has this announcement taken so long? (Score:4, Insightful)
There's a high-stakes poker game being played here. They don't want to show their hand too early.
If they don't go after WebM, then eventually the H264 licensing falls apart. If they do go after WebM, there will be a challenge to the patents, and they potentially lose everything. So there's some delay and some bluffing going on.
Re: (Score:2)
Re: (Score:2)
You say this now. But this is about the long haul. There is already hardware that supports WebM.
Because most people buy their smartphones with a plan, replacement rates are high.
The numbers say: after 3 years of use, 82.6% will have been replaced (and 18% after the first year).
So if you add it to every Android phone today and the Android phones remain populair, in 3 years almost all Android phones people are using support it in hardware.
Re: (Score:2)
I would suggest that the sheer number of videos already encoded in H.264, and the number of devices that support it already is a significant advantage. WebM still doesn't have the same quality as H.264 given the same bandwidth either. Therefore, I would say the only real advantage that WebM has is that it is mostly/sort of/maybe royalty free, which isn't much of an advantage when the royalty is less than $.001 per device.
Re: (Score:2)
The sheer number of videos encoded in Real's format sure gave them a significant advantage, no? No?
The royalty issue is for broadcasters and production houses, not 'devices.' Why the hell should I pay MPEGLA licenses for an independently shot movie I want to release on the web in a commercial fashion? What have they created with H264 that is novel and 'unique' other than squeezing more pixels into a smaller file size that justifies paying the royalties to use technology that is a glorified zip archive?
Re: (Score:2)
The number of videos encoded in Real's format was tiny to begin with, and none (virtually none) that were professionally done. There is quite a difference. I would venture to say that the majority of people out there have at least 1 H.264 video now, whether it be from a bluray, a downloaded video, a home made movie from a video camera, or a cell phone clip. That's magnitudes higher penetration that real ever was.
Actually, royalties are paid per device, but there is a cap and it's so low that most broadca
Re: (Score:2)
People seem to forget that the one and only thing that matters when looking at WebM adoption is the HTML5 video tag and hardware (which exists).
Re: (Score:2)
Google is not the arbiter of "open" (Score:1, Insightful)
Google has zero credibility when it comes to promises of "open standards." They've hung their entire Android ecosystem out to dry and abandoned its partners to fend for themselves against patent holders.
WebM is most likely toxic and has an uncertain future. Google knows this and that is why they don't offer any type of indemnification to parties who choose to bet on the codec.
This "Google says it's open, therefore it's open" nonsense has to end.
Re:Google is not the arbiter of "open" (Score:5, Insightful)
that is why they don't offer any type of indemnification to parties who choose to bet on the codec.
No, they don't offer any type of indemnification because patent trolls file bullshit patents like claiming a "buy now" icon in your app infringes on a patent regarding two-way customer feedback. At this rate, whoever has that patent on amusing a cat with a laser pointer has a solid case against WebM. It's impossible to defend against all the bullshit.
If the patent system wants to continue down this road, then the patent laws need to be revised that if anyone brings a patent case against someone, and they're found to not infringe on the patent, the patent holder owes double the defense's court fees and lawyer costs. Triple if the plaintiff tries to bail out after a Markman Hearing told them their patent doesn't mean what they want it to mean, and there's no way in hell that it's being infringed on.
Re: (Score:2)
We need a more "European" system.. Once something is brought before Court, you should not be able to "settle". In both Civil and Criminal court too many cases are known to be losing by a bigger party, just to drain resources from a smaller one.
We need to adjust the system to the "Court's Satisfaction". In a criminal case like OJ or Casey, the resulting perjury and damage charges for losing the case should all be presided by the same court... So the court can punish the actions of it's agents, not just the p
Re: (Score:2)
'We need a more "European" system' where software patents don't apply. Or atleast a whole lot less, there are always exceptions.
Re: (Score:2, Insightful)
Neither does MPEG-LA
Re: (Score:2)
Prepare to get modbombed for criticizing Google.
No, only for spreading old and stale FUD against Google.
Meanwhile, the people voting you down won't address the fact that Google provides no indemnification.
Obviously, since you can't post and mod at the same time. Other people however have addressed the point many times in the past: you don't get patent indemnification *anywhere* else in the industry, and that's because large corporations don't remain large by being naive enough to think that judges are infallible and would never avail the insane demands of a patent troll.
how about a less-efficient free codec? (Score:1)
If Google and others were to buy out or develop patent-free less-efficient codecs that get the job done, AND push them in preference to non-free codecs, we could have the benefits of widely-used free codecs, albeit with a bit of inefficiency.
Re: (Score:2)
And if Google did do that there would be people who would cry and complain for entirely different reasons. Google knows that they can't please all the people all of the time, and I'm pretty sure they are ok with that.
Re: (Score:2)
And if Google did do that there would be people who would cry and complain for entirely different reasons. Google knows that they can't please all the people all of the time but most of them will use the product anyhow, and I'm pretty sure they are ok with that.
FTFY
Re: (Score:2)
Well yes, that was sort of implied as Google doesn't keep underperforming products around long.
Re:how about a less-efficient free codec? (Score:5, Insightful)
And let's please remember (Score:5, Interesting)
Google checked all this out. Before they bought On2, they got to see all their IP, all their sutff (under NDA of course) as is standard practice for buyouts. Also, after they bought them, they took some time to turn VP8 in to WebM. During that time you really think they didn't do a through checking of patents that might apply? Also consider that Google is the best of the best at searching and data mining. They probably found everything.
My bet is they concluded that any or all of the following are true:
1) WebM does not infringe on any of the legit video patents out there.
2) That any patent WebM does infringe on is one that can be showed to be invalid via prior art.
3) That anyone who has a valid patent, Google has a more damaging counter patent(s) and thus they'll have to back down.
I cannot believe that Google ran in to this without doing good research. I also find it easy to believe that MPEG-LA is grasping at straws, particularly given how long it has taken and the lack of specifics.
Re:And let's please remember (Score:4, Interesting)
WebM does not infringe on any of the legit video patents out there.
A company with the caliber of On2 to hold patents that do not infringe on h264 and still claim to reach h264 quality is hard to believe.
That any patent WebM does infringe on is one that can be showed to be invalid via prior art.
Video compression is a mature area and you have to fight teeth and nails get your IP in the standard (I attended the VCEG/MPEG standardization meetings for h264, I witnessed the blood first hand). The IP holders are huge companies http://en.wikipedia.org/wiki/MPEG_LA [wikipedia.org], and they have very decent research labs.These guys are not fucking around, they do their patent research before fighting to push their IP in.
That anyone who has a valid patent, Google has a more damaging counter patent(s) and thus they'll have to back down.
That sounds like pure fantasy.
I also find it easy to believe that MPEG-LA is grasping at straws, particularly given how long it has taken and the lack of specifics.
MPEG-LA does not seem to know what to do with h264 either. They keep pushing the "end of free license" period. We are safe until 2016 for now I believe. That is about when h265 should be finalized.
Re: (Score:2)
MPEG-LA does not seem to know what to do with h264 either
There are 1,027 AVC/H.264 Licensees [mpegla.com]
Digital television. Broadcast, cable and satellite distribution. Big content. Disney. Blu-Ray. Netflix. Medical imaging. Industrial and militray applications.
Someone seems to have a clue.
While WebM remains little more than mediocre YouTube transcode playable in a browser.
Re: (Score:2)
Re: (Score:2)
Apple devices have big enough margins to easily absorb proprietary codec royalty fees for H264/AAC whereas Android phones have much smaller margins and this is why Google want a free video codec
The "Enterprise Cap" on H.264 royalties is $6.5 million a year.
The maximum is 20 cents a unit on sales of 100,001 to 5 million units. 10 cents a unit above 5 million units. SUMMARY OF AVC/H.264 LICENSE TERMS [mpegla.com]
These charges are spread across your entire product line.
Re: (Score:3)
This video, shot at Mozilla, explains about the origin of Ogg Theora (which was based on VP3) and thus also WebM (which was based on VP8):
http://air.mozilla.org/open-video-codec-discussion-at-mozilla/ [mozilla.org]
Re: (Score:2)
On2 were developing codecs long before H.264 was developed. On2 have never been an MPEG-LA member: it is equally possible that On2 hold patents that H.264 infringe upon.
Re: (Score:3)
Just read again what you just wrote. A mature area... fight teeth and nails to get your IP into the st
Re: (Score:2)
MPEG-LA does not seem to know what to do with h264 either. They keep pushing the "end of free license" period. We are safe until 2016 for now I believe. That is about when h265 should be finalized.
They're still collecting plenty cash from all the people who want to produce H.264 content. They're just checking to see if they dominate the market so completely they can start double or triple-dipping, making people pay for both producing and streaming and viewing. The extensions mean "not yet", maybe they'll continue all the way through the last patent - I hear 2027 - but don't bet on it.
Re: (Score:2)
During that time you really think they didn't do a through checking of patents that might apply?
Two possibilities: 1) Google is huge and arrogant and maybe they didn't do a great job at this and/or 2) given the complexity of modern video codecs, it is tough to do a good job of working this out, especially given that VP8's definition has mainly been source code.
Re: (Score:3)
Remember: whether there exists a real patent that actually does predate whatever Google is doing with WebM has no bearing on whether lawsuits will be filed. See also http://www.thisamericanlife.org/radio-archives/episode/441/when-patents-attack [thisamericanlife.org].
Re: (Score:2)
During that time you really think they didn't do a through checking of patents that might apply?
Oh, absolutely. For two reasons. First, because Google has a long track record of completely ignoring anyone else's IP. Second, because doing such a search is REALLY hard. What do you search for? Break the algorithm down into small steps and then see if each is patented? But patents can't be easily searched for algorithms - they don't, for example, require a pseudocode representation of the algorithm in a consistent format. You're taking a bit of code, and then searching terabytes of text to see if a
Re: (Score:2)
3) That anyone who has a valid patent, Google has a more damaging counter patent(s) and thus they'll have to back down.
This is where GPL v3.0 enters the game. Most people are quite Ok with the h.264 patents, except that you can't write a codec that is GPL v3.0 licensed, because you can't have GPL v.3.0 code that implements a patented invention, unless you have a patent license that covers the code and any GPL licensed copies or derivatives.
If someone has a valid patent that WebM is infringing on, and is convinced by threats from Google not to sue, that doesn't make a difference - the code would be infringing and cannot b
Re: (Score:2)
But this is the point... Google did build this from what they believed was a patent-free, or at least patents they thought they has rights to... But MPEG-LA is such a massive licensing firm, they'll FIND somebody in their group with something Google could possibly infringe on. The group has gone from being a way for big companies to AVOID lawsuits, to a group of old guys that beats up new kids with 20-year-old crap.
Google is of the "don't ask, charge ahead" approach to patents. Google is throwing CS Masters
Re: (Score:2)
All the patent pool members are scared to get one of their gravy train patents to be ruled invalid, finding one without some below the table (illegal) pay off to to make the risk of invalidation worth taking is not at all easy.
Re: (Score:2)
The problem is, the only codecs that are pretty much guaranteed to be free are early versions of MPEG1, and MJPEG.
Both of which, in 2011, are downright awful codecs.
(Although, my digital camera shoots MJPEG...)
Re: (Score:2)
OK, for the purposes that most people use video codecs for in 2011, those are awful codecs - either filesize is absolutely huge for what's considered reasonable quality now, or image quality is downright awful for what's considered reasonable size now.
Speaking of WebM (Score:2)
When is Google pulling h.264 support out of Chrome? I just did a new install on a previously Chrome-less Windows box, and it handled an h.264-only <video> element just fine.
I knew the Mac version of Chrome still played h.264, but I'm not sure if it's relying on its own internals or on the OS in that case.
Windows 7 has h.264 playback built in (Score:2)
And previous versions of Windows can have it added with codec packs. Chrome could well use the media layer in Windows for playback of any formats it does not support internally (it would make sense to do so).
Re: (Score:2)
I did the test on an XP box that did not previously have Chrome installed - the h.264 video played just fine.
And yes, I know it was h.264 because I put both the video and the page together.
Re: (Score:2)
h.264 hasn't been removed from Chrome yet. So it is really nice you tested it, but it was to early.
Is it a codec or a fake antivirus? (Score:2)
And previous versions of Windows can have it added with codec packs.
How many fake antivirus "products" masquerade as codec packs?
Chrome could well use the media layer in Windows for playback of any formats it does not support internally (it would make sense to do so).
That would have two drawbacks:
I was referring to the installer (Score:2)
Re: (Score:2)
When is Google pulling h.264 support out of Chrome? I just did a new install on a previously Chrome-less Windows box, and it handled an h.264-only <video> element just fine.
I knew the Mac version of Chrome still played h.264, but I'm not sure if it's relying on its own internals or on the OS in that case.
During the rise of the WebM debate Google yanked H.264 from Chrome for Linux and a few months later added it back. They have continued it, but don't include MPEG-4 support.
Chrome version: 14.0.835.8 dev:
This has been supported for at least the past 6 months.
Re: (Score:2)
Probably after Chrome 13, AFAIK.
Maybe more importantly, when will Youtube automatically use HTML5/WebM when the browser supports it.
As I understand it, all new uploads are already converted WebM.
3 months ago Youtube had already converted 30% of the videos, which make up 99% of all views.
At the last Google I/O someone working on the new Youtube 'player' mentioned embedded worked and it looks like they are now very close to releasing it.
Re: (Score:2)
Chromium not Chrome. Chromium is the open source version, which comes without things like built in pdf readers, h.264 support and built in flash.
While I realize the post was in the Chromium Blog, the January blog post [chromium.org] says h.264 support is going to be removed from Chrome "to make it consistent with the codecs already supported by the open Chromium project."
No mention of Apple? (Score:2, Informative)
Why no mention of Apple? Or Microsoft? They are both members of MPEG-LA [wikipedia.org], not to mention other sleazy organizations [wikipedia.org].
Re: (Score:3)
Why no mention of Apple? Or Microsoft? They are both members of MPEG-LA, not to mention other sleazy organizations.
There are about thirty AVC/H.264 licensors, most of them global industrial giants like Mitsubishi, Philips and Toshiba. AVC/H.264 Licensors [mpegla.com]
There are about 1,030 H.264 licensees. AVC/H.264 Licensees [mpegla.com]
H.264 is theatrical production, Blu Ray, broadcast, cable and sattelite distribution. Medical, industrial and military applications. There is no such thing as an HDTV set or set top box that doesn't support H.264.
WebM is, well, WebM.
Google is an H.264 licensee. Google is hedging its bets.
Re: (Score:2)
Why not indeed ? People like to paint Google as some kind of altruistic entity but look at who they are taking on here, their direct competitors. Google are the Microsoft of the new millennium: they have a very profitable core business in search and advertising that could possibly be threatened from developments in other markets so they take on all those companies and go into the phone business, the browser business, the netbook business, social networking, TV streaming, etc. All to keep others from having
Re: (Score:2)
the sad thing about android is that due to numerous patent licensing issues, no new manufacturer can actually enter the market with it. it's all just the same old companies and their few chosen models, all aping each other in form and function. no innovation, no nothing.
sure, anyone can order chips that run it, whole 3g capable socs. but you can only sell them in low numbers or you'll be spanked by patent police and you'll end up paying 60 bucks or more for the licenses per phone. that's a LOT when in two y
Who does this surprise? (Score:2, Interesting)
Who does this surprise? And I'm not talking about patent trolls or other attacks just looking to make a quick buck. I mean who is surprised that WebM steps all over patents associated with h.264 and other video standards? Anyone?
The one and only argument I have ever seen in favor of WebM is that it doesn't have licensing restrictions. That's it. From a code point of view h.264 is at least as open source seeing as many of the best tools and compressors are (and always have been) open and in many cases free.
Re:Who does this surprise? (Score:5, Insightful)
who is surprised that WebM steps all over patents associated with h.264
Objection. The correct sentence here is "who is surprised that MPEG-LA claims that WebM steps all over patents controlled by MPEG-LA".
Given that WebM was specifically designed to not infringe any patents, I for one would be very surprised if it "steps all over" a large number of patents.
Sadly, I would not be surprised if Google lost a lawsuit over some patent that someone claims covers WebM. But that's because there are so many patents, and stuff happens in lawsuits.
The one and only argument I have ever seen in favor of WebM is that it doesn't have licensing restrictions.
Isn't that enough?
The basic pitch for WebM is "not as good as H.264, but you are free to use it". If you need the best possible video encoder and are willing to abide the the licensing restrictions and fee schedules of H.264, you use H.264. If you are Debian, and you only ship free software that isn't patent-encumbered, you ship WebM and not H.264. This isn't rocket surgery.
From a code point of view h.264 is at least as open source seeing as many of the best tools and compressors are (and always have been) open and in many cases free. WebM doesn't do things significantly different from a technical point of view or an implementation point of view. It is substantially the same technology.
Your point about H.264 being available in open source is accurate, but pointless. Your second point is correct if we agree that the word "substantially" covers a lot of differences. The differences make H.264 the better encoder.
So if you have two projects that are effectively identical, but one has licensing restrictions on some streaming content and the other doesn't. Why would patents that cover one not cover the other?
Because patents aren't judged by this "substantially" word you used. Patents cover specific things. On2 seems to have studied patents to figure out what they couldn't do, and found ways to do things that work almost as well without being covered by the patents.
The dangers here are that On2 overlooked a patent or otherwise made a mistake; or that a court would rule that On2 skated too close to the line. There is no danger that a court will say "WebM is substantially doing the same thing as H.264, so all the H.264 patents apply." You didn't really mean to imply the situation was that simple, did you?
This stuff is like the JPG and GIF patents. Made zero real world impact and by the time everyone was finished arguing over them the patents had already expired.
There is one major difference: the patents on H.264 are not just about to expire. I did a quick Google search and found that you would have to wait until 2025 to use H.264 for free.
If you use a patented format, you have to get a patent license. The owners of the patent license get to dictate the terms of the license.
Google, who owns and runs YouTube, doesn't want to build its business around H.264, because then when it is time to renew the licenses, the terms or fees could become Draconian.
Google doesn't want to build YouTube around an old, lousy video coder that happens to be free, because the users won't be happy their videos load slowly and look horrible; and Google has to pay for the bandwidth.
So: WebM. Not as good as H.264, but nobody can use it to tell Google "you now have to give us big large huge royalties on your use of this video coder". They can predict their future licensing costs (zero), and their bandwidth costs aren't horrible and the user experience is good.
WebM benefits everyone except for the people who own H.264 patents. And they will still make money on H.264; they just won't be the only game in town anymore. They want very much to be the only game in town and charge whatever they feel like charging. I don't understand why you are so keen on this idea.
steveha
Re:Who does this surprise? (Score:4, Interesting)
Given that WebM was specifically designed to not infringe any patents, I for one would be very surprised if it "steps all over" a large number of patents.
WebM was not developed within an ISO recognized standards development organization. It never had a hope of being patent-free: the process was not open, not due-process, and for that reason and others the folks who might have been willing to donate their IP to the effort were not properly motivated.
There is one, and only one video codec that was developed in an open standards process specifically to be non-royalty, and that is JPEG 2000 Part 1 aka ISO/IEC 15444-1. During the development process of JPEG 2000 Part 1, an agreement was reached with over 20 large organizations holding many patents to allow use of their intellectual property in connection with JPEG 2000 without payment of license fees or royalties.
That is a major reason why JPEG 2000 is the basis of Digital Cinema.
Re:Who does this surprise? (Score:4, Interesting)
WebM was not developed within an ISO recognized standards development organization. It never had a hope of being patent-free: the process was not open, not due-process, and for that reason and others the folks who might have been willing to donate their IP to the effort were not properly motivated.
Neither was Vorbis. Vorbis has been used in huge projects by huge companies for over a decade. Why has it remained royalty-free? Why doesn't VP8 similarly have "a hope" of remaining royalty-free? Until the MPEG LA or one of the companies that allegedly hold patents essential to VP8 have something substantive to show there is no proof that VP8 infringes any patents. All the MPEG LA is doing now is what they've always done since the launch of WebM: scaremongering.
Re: (Score:2)
WebM was not developed within an ISO recognized standards development organization. It never had a hope of being patent-free: the process was not open, not due-process, and for that reason and others the folks who might have been willing to donate their IP to the effort were not properly motivated.
On the contrary, that's why WebM stood a chance. ISO are generally not interested in royalty-free; they're generally quite happy for members to inject requirements into standards that require everyone to license their patents. (Also, remember that very few people use JPEG 2000 - that's probably the only reason we haven't seen massive patent lawsuits over it it. Wavelets are much more of a patent minefield than DCT-based codecs.)
Re: (Score:2)
Given that WebM was specifically designed to not infringe any patents, I for one would be very surprised if it "steps all over" a large number of patents.
Except that WebM wasn't designed to be patent free. It was designed to be license free. Neither WebM or the related VP8 or Matroska projects were ever designed to be patent free. They were all built around free licenses, not patents. Not the same thing at all.
Under current rules, I doubt that it is possible to create an image or video format that is patent free. Doesn't mean I am defending the current situation, but it is the reality on the ground at the moment, and likely for the foreseeable future.
WebM is
Re: (Score:2)
WebM wasn't designed to be patent free.
Very well, then: WebM was designed to be free of patents other than the On2 patents that Google owned and freely licensed to allow the use of WebM.
Does the above statement meet your required level of precision?
Neither WebM or the related VP8 or Matroska projects were ever designed to be patent free. They were all built around free licenses, not patents. Not the same thing at all.
If any software uses patented technology, the patent owner can control the use of that soft
Re: (Score:2)
Except that WebM wasn't designed to be patent free. It was designed to be license free
VP8 was designed to be patent free. On2 took H.264 and then rewrote all of the bits that were patented to be subtly different in the hope of avoiding any patents, except for the ones that they owned. I don't know what you mean by 'license free' - VP8 is covered by two licenses, a copyright license on the source code and a patent license. Both grant royalty free use of the On2 code and patents that Google purchased.
Re: (Score:2)
Objection. The correct sentence here is "who is surprised that MPEG-LA claims that WebM steps all over patents controlled by MPEG-LA".
Others [multimedia.cx] have already predicted VP8 would run into patent issues.
Anyone who knows even the least bit about video codec technology can tell you that it's virtually impossible to design an advanced video codec that does not infringe on any video coding patents, which isn't surprising if you consider the technology involved is decidedly non-trivial, and took over 3 decades to develop. H264 is currently the most advanced standard for video coding, but it didn't materialize out of thin air either: it simply build
Re: (Score:2)
So Google and On2 know nothing about video codec technology?
Re: (Score:2)
Video codec patents arent simply about the nice mathematics. They are about efficient ways of performing that math on various technologies. Thats the real situation.
It isn't that the overall math in play can't be accomplished in ways that avoid patents, its that its very hard to do any of it nearly as efficiently as H.264 with
Re: (Score:2)
He didn't even read the patents ... he compared features.
He is relatively young and way too busy to try to examine the research background for each of those features. He should never have judged something as validly patented in the H.264 patent pool just because he saw it first in H.264 ... it was a stupid thing to do (look at the intra prediction redaction) and a misuse of his own reputation.
Doctrine of equivalents (Score:3)
patents aren't judged by this "substantially" word you used
"Substantially" in patentese is called the doctrine of equivalents [wikipedia.org]. Perhaps that's what you were thinking of when you mentioned the possibility "that On2 skated too close to the line."
Re: (Score:2)
New rule (Score:5, Insightful)
They should make a new rule. If you don't immediately(reasonable amount of time, 1-2 months?) sue another company once you found out they've made a product based on your patent, you give up any rights to said patent.
You shouldn't be able to sit on it and wait for it to be more "lucrative" to sue.
Re:New rule (Score:5, Informative)
You shouldn't be able to sit on it and wait for it to be more "lucrative" to sue.
I am not a lawyer, but I believe this is what you wished for; it's already part of the law.
http://en.wikipedia.org/wiki/Laches_(equity) [wikipedia.org]
steveha
Burden of proof (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
They should make a new rule. If you don't immediately(reasonable amount of time, 1-2 months?) sue another company once you found out they've made a product based on your patent, you give up any rights to said patent.
You shouldn't be able to sit on it and wait for it to be more "lucrative" to sue.
That's essentially what the doctrine of Laches [wikipedia.org] (a form of Estoppel [wikipedia.org] is all about. The plaintiff knew about a potential patent violation but sat on it for some length of time until the defendant actually committed themselves to a position where a lawsuit would be more harmful.
This is going to be part of Google's case against Oracle [groklaw.net] after having uncovered Jonathan Schwartz's blog entry congratulating Google on using Java technology.
It's a tough case to argue but it can be done.
And that is how patents are promoting progress. (Score:3)
And that is how patents are promoting progress. A company makes a clean room afford to come up with a new algorithm and tries in it's best to identify and not use any patents, and another company buys it and is releasing the algorithm under a free license to improve the life of everyone.
The only way to advance in the field of video codes is a) be the lucky company which is in the patent pool or b) wait until all patents expire. How is that suppose to promote the technology again?
Re: (Score:2)
A company makes a clean room afford to come up with a new algorithm
I'm assuming you mean effort when you say afford, but even then it's wrong. On2 did not make a clean room effort. VP8 is H.264 + some tweaks to avoid patents. They took the research of others and made some minor adjustments to avoid paying royalties. Whether you think the people doing the research should be compensated, or whether patents are a good form of compensating them, is still an issue, but let's not pretend that the On2 engineers sat on their own without access to the H.264 and worked out a com
Re: (Score:2)
Since VP8 is somewhat older than H.264 I find that chain of events vert hard to believe... In fact I considered it logically impossible.
More likely both formats are based on the same research. Real research done by real researchers, who usually work for universities and publish their research royalty free to meet their published articles count, and to advance their academic career. Just like 99% of all genuine advancements in computer science are made.
Re: (Score:2)
Since VP8 is somewhat older than H.264 I find that chain of events vert hard to believe... In fact I considered it logically impossible.
Really? From Wikipedia [wikimedia.org]:
The standardization of the first version of H.264/AVC was completed in May 2003
Also from Wikipedia [wikimedia.org]:
The development of the codec[VP8] was announced by On2 Technologies on September 13, 2008 to replace its predecessor, VP7.
And again [wikimedia.org]:
On2 Technologies announced TrueMotion VP7 in January 2005.[1] The public release of VP7 codec software was available in March 2005
And going further [wikimedia.org]:
The VP6 codec was introduced in May 2003.[1] In October 2003, On2 officially released its TrueMotion VP6 codec.[2]
So, VP6 was developed at the same time the final H.264 standard was published. VP7 and VP8 were developed years later.
Re: (Score:2)
Some parts of H.264, especially the parts where it is better than VP8 are very very recent. H.264 is actually not just one standard. In the same way VP8 is the latest iteration of VP6.
So compare VP6 to early H.264 or compare VP8 to equivalent recent H.264 versions. Even in the best cases they only come out even.
Re: (Score:3)
Re:Who is Florian Mueller? (Score:5, Insightful)
A professional troll, who keeps posting hatchet-jobs stories about Google/Android only to have them promptly rebuffed.
Re: (Score:3)
Re: (Score:2)
The companies* involved in the W3C could not agree on a video codec, I still wonder if Google had completed their aqusition of On2 before the W3C discussions, would it have mattered ?
Actually the IMG-tag also does not have anything in the specifications about BMP, XPM, JPEG, GIF or PNG so it isn't all that strange.
* Of which some are obviously members of the MPEG LA, like Apple and Microsoft if I remember correctly.
Re: (Score:2)