Google and MPEG LA Announce Agreement Covering VP8 Video Format [Archive]

paradoxical

7th March 2013, 21:02

That's a good thing to hear. Guess that definitely disproves the notion that VP8 was not patent-encumbered. Also, that the people raising legitimate warnings that the format was not patent-unencumbered were definitely not FUD as so many were quick to claim that they were. Too bad other video/audio formats can't have such a similar licensing agreement, though. Those are quite generous terms compared to others.

hellfred

7th March 2013, 21:32

Maybe MPAG LA came to the conclusion that risking a court trial might earn them the revocation of multiple of the patents, which are the base and fundament of their business.
(See Google vs. Oracle trial over the Java patents)
Maybe it was therefore some kind of draw.
But all of that are mere speculations.

paradoxical

7th March 2013, 22:01

Possibly, but I would imagine that if the patents in question were so weak that Google would not have just rolled over and settled. It would be far better for them to get the patents ruled invalid or the VP* formats found non-infringing so as not to have had to license anything. The very fact that they fought the Oracle v. Google case so hard and would actually go contrary to your claim.

hajj_3

8th March 2013, 00:45

It is nice to know that people don't need to worry about patents for vp8 anymore. Even more exciting is that VP9 has licenced those patents too so unless they use some other patents that they haven't just agreed to licence then vp9 should have no patent problems either which is great. Mozilla won't have this ease with their Daala codec as they don't have the money or commercial hardware pressure that google has. Lets just hope mozilla/xiph and google join together to create vp9 instead of seperate codecs.

iwod

8th March 2013, 12:13

Well VP8 is old news, I hope the same could be done on VP9 or VPx. And hopefully xvp8 is still in the work.

Kurtnoise

8th March 2013, 12:22

The agreement allows Google to sub-license the techniques covered by the agreement to any VP8 user and also covers the next generation of the VPx codec.
http://techcrunch.com/2013/03/07/google-and-mpeg-la-sign-licensing-agreement-covering-googles-vp8-video-codec-clearing-the-way-for-wider-adoption

mandarinka

8th March 2013, 18:55

There is a question though - is VP9 covered only for VP8 techniques, or is there also a promise to not sue about new techniques that might infringe on additional patents?

When that proto-pool (12 companies, now just 11) was being formed, there was no sign of VP9, so there might be subjects that weren't taken in because they didn't have "stuff against VP8", while they may at they same time "stuff against VP9" and in theory could be free to sue.

There is no way to tell how it is from the press release, but I am hoping some information on that will be released later. It would be nice for Google if VP9 was "approved" generally by MPEG-LA as part of the agreement.

mandarinka

14th March 2013, 04:30

http://www.pcper.com/news/General-Tech/WebM-Survived-MPEG-LA-Nokia-Right-Back-Down

Now, Nokia is suing HTC over two VP8-related (allegedly) patents, infringed by Android phones. Nokia isn't part of the MPEG-LA deal, and thus is free to sue (and keep the patent uncertainty about VP8 up).
There is some discussion on IETF mailing list (http://www.ietf.org/mail-archive/web/rtcweb/current/threads.html#06704) about this, but there is little public info yet it seems.

Hopefully Google settles with Nokia too... it might attract more litigators, but I doubt there will be more of Nokia's calibre.

oibaf

14th March 2013, 14:53

Onward! - Now that the distractions of the MPEG LA licensing initiative are behind us, it's a good time to review recent improvements with VP8 and take stock of VP8 adoption and proliferation (http://blog.webmproject.org/2013/03/onward.html)

Drop H.264 from YouTube and Chrome?

paradoxical

14th March 2013, 14:59

Drop H.264 from YouTube and Chrome?

Hahahaha. That's a good one. So what is the fallback going to be for the nearly 1 billion phones that have only hardware H.264 decoding?

Oh and I had to chuckle at:

We realize that dramatic headlines attract readers, but despite some of the speculation, there was never any lawsuit, and there was certainly no "finding" or "admission" of infringement.

So if there was no finding of infringement and they weren't guilty of infringing anything why didn't they just allow themselves to get sued and get the patents invalidated and/or the codec shown to be non-infringing in court? They didn't seem to have a problem doing so in Oracle v. Google with Android and Dalvik. Sounds like spinmeistering to me. If anything, settling with MPEG-LA only makes them a bigger target.

mandarinka

14th March 2013, 17:20

Yeah, naturally they need to do damage control + PR spin. I think they might have been pressed by circumstances. H.265 is almost done (but VP8 hasn't made a breakthrough yet - it's chances will therefore only get lower as time passes), they want to get VP8 rubber-stamped in MPEG and at teh same time, they propose it as mandatory codec for webrtc. In the overal strategy, it probably seemed advantageous to send a signal to those various parties that VP8 is safe. Maybe the trial with Nokia also contributed to the pressure - if they faced legal threats from two sides at once, it would look much more bad IMHO.

dapperdan

14th March 2013, 17:33

Actually getting sued over a specfic claim of a specific patent seems like pretty good progress for royalty-free codecs. In the bad old days you just had to tell campfire ghost stories about the legendary number of multimedia patents to stop progress in its tracks. And those were much harder to disprove. Now we're getting some actual meat. I'm no expert but the two Nokia patents seem just as BS as any other patent that gets thrown around in these types of fights.

Any thoughts on why Nokia is suing HTC rather than the deep pocketed inventor and biggest distributor of VP8 encoders, decoders and VP8 encoded material?

edited to add: my money is on Nokia going full troll within a year.

mandarinka

14th March 2013, 20:36

HTC is their direct competitor. Also naturally first you go after a smaller player, the suit will be a bit smaller-scaler (less hazardous), and if successful, it is a stepping stone for the next trials.

As for this being good, I can't agree. Also, if it was up to "the internet", every patent is BS, obvious (and prior-arted at the same time) and should not have been granted in the first place. But generally internet is clueless - this should be left to people who actually understand the area.

Edit: also, as you can see, Google's aim in these affairs is to get a settlement and buy perpetual sublicenseable rights to the VP8 patents in exchange for some payment.
To maximise gains from such deal, Nokia needs to put pressure on Google, the more the more money they can whack out of it if infringement is found/ Google looses nerve.
Pressuring Google's client (android vendors) is a good way towards that.

paradoxical

23rd March 2013, 23:48

The list of patents Nokia are asserting against VP8 are here (http://www.fosspatents.com/2013/03/setback-for-googles-vp8-nokia-refuses.html#nokiavp8iprlist). And they are not wanting to license them under either royalty-free or FRAND terms. Looks like Google just painted a big target on itself with its PR blustering.

xooyoozoo

24th March 2013, 00:43

And they are not wanting to license them under either royalty-free or FRAND terms

They're refusing to license under any terms, if I'm reading it right.

I can't really tell what they'd be gaining from a hardline stance, as most of the peripheral companies associated with VP8 would just shrug and move on. I highly doubt VP8 is a critical part of any corporate infrastructure.

Google would lose a bit of face, sure, but they're no strangers to dropping dying projects, and that won't get Nokia any money.

mandarinka

24th March 2013, 03:35

I don't think they categorically refuse licensing - they just refuse any forms of "FRAND", much less royalty-free.

What that means is that Nokia says it reserves the right to offer licenses at any terms that they would prefer at a time - for example, to charge excessive sums, to deny a license to gain a competetive advantage against somebody, or to be able to use the patents at a court as an "extortion device", for example to demand an import ban (which is the big strategic victory that all these companies seek in a patent war - if you get that, you have a huge edge to force a cross-settlement on favourable terms). If you pledge to license on FRAND terms, you are barred from using those patents like that.

In most cases, they will give a license, because in the end, they want to capitalise on the IP. But in cases where a higher gain could be get from refusing a license, they will refuse it to get that sweeter gain...

Basically, they say they consider the IP valuable and they warn that they don't intend to limit its value by limiting it with a FRAND pledge. It is their patents, they didn't come with the VP8 thing, so from the PoV of their economic interests, you can't really blame them. World isn't a place where all goes without problems and conflicts of interests...

Edit: there is still a way out of this (I assume Google won't be able to invalidate/prove noninfringing for all the patents). I think that Nokia understands that Google wants to keep the royalty-free scheme and without that VP8 probably isn't interesting for them. So they probably just want to get as much money as possible as part of a settlement similar to MPEG-LA's one (one time or perpetual fee...) and eventually will agree to it - it just might cost Google a lot of money. It still would be a good outcome for the users if it happened.

hajj_3

24th March 2013, 23:03

I just hope that google either pays nokia a wad of cash to get it licenced; takes them to court and tries to invalidate each patent or they change vp9 so that it doesn't violate their patents and forget about vp8.

dapperdan

25th March 2013, 12:01

A slightly better list I found for the Nokia patents, with links to the relevant patent info:

http://mdpaste.appspot.com/p/agdtZHBhc3Rlcg0LEgVQYXN0ZRjJoxYM

Because of the way patents get written they start off quite broadly then narrow right down to incredibly specific, so while they do make broad claims on things that VP8 clearly does (adaptive loop-filtering, alt-ref frames) it's hard to tell exactly how identical the implementation needs to be before it'll stand up in court (and relatedly, how easy it would be to vary the implementation to work around the patent, as seen recently in Apple vs Samsung UI patent cases).

I just read another one at random, and it actually has the stereotypical bad patent joke "X on a mobile" as the last claim:

Filtering method, filter and mobile terminal (http://worldwide.espacenet.com/publicationDetails/claims;jsessionid=86F41CA26E9F63BDCF2A43EF797B9158.espacenet_levelx_prod_0?CC=EP&NR=0884911A1&KC=A1&FT=D&ND=&date=19981216&DB=&locale=en_EP)

7. A terminal device (90, 100) according to the claim 6, characterized in that it comprises receiving means (91, 92) for receiving the video picture via mobile communication network.

How does that even make sense? If you've got a solid patent on some algorithm, then they can't use it on any kind of device. And if you don't, then why does adding that line help you?

paradoxical

25th March 2013, 15:26

I don't think they categorically refuse licensing - they just refuse any forms of "FRAND", much less royalty-free.

Actually his reading sounds way more correct in light of quotes such as:

As a result, we have taken the unusual step of declaring to the Internet Engineering Task Force that we are not prepared to license any Nokia patents which may be needed to implement its RFC6386 specification for VP8, or for derivative codecs.

Emphasis added. Basically their reasoning is that VP8 is Google trying to force a standard from an inferior, (previously) proprietary codec rather than collaborating with other industry groups to make a standard like with H.264, etc.

JEEB

25th March 2013, 16:29

I would have not thought I'd be agreeing with FOSSPatents on anything lately, but I think I just spewed my coffee for finding some relatively sound thoughts in a blog entry (http://www.fosspatents.com/2013/03/nokia-comments-on-vp8-patent.html) about Nokia's comments on the anti-VP8 stance at the IETF, and ending up generally agreeing with the fact that we should think about certain things a bit more. As someone who doesn't like software patents I am not sure if I like how this is done (although, to be honest, were there other ways of bringing this point up as strongly?), but I do see some points that are worth raising amidst all this.

Now, the blog entry does contain some general critical points towards Google in general, but if you leave those out there are quite sound points left in there that are (unfortunately) often left out of the VP8/9 discussion. And quite unfortunately I haven't (yet) seen anything different to how VP8 was made in how VP9 has been made so far (other than we can look from the side as the reference implementation is being built source code wise).

paradoxical

25th March 2013, 17:09

Yeah, the tactics of Nokia are definitely debatable but their points against VP8 and Google are mostly spot on. Google is basically using their weight in the industry to hamfist a one-party standard on the web and only at the end attempting to get some form of rubber-stamping from standards bodies. If these were such great standards then they should live and die by their merits, not because they have a backer with deep pockets and industry weight.

mandarinka

25th March 2013, 18:04

Well, if they actually went friendly (FRANDly heh) and generously gave away the rights, then Nokia would in fact sort of neglect to protect the value of its assets. Since they aren't doing that well financially, I don't think they can be criticised for being greedy.

mzso

26th March 2013, 12:59

I wonder if a year from now we'll be reading about HEVC and VP9 also drowning in patenting crap...

paradoxical

26th March 2013, 13:31

I wonder if a year from now we'll be reading about HEVC and VP9 also drowning in patenting crap...

HEVC should be no different than any other MPEG/ITU video standard. It'll have a patent pool with MPEG-LA, etc. and everyone implementing the standard knows the score with respect to patents. VP9 is likely going to see the same issues as VP8 is seeing now. That's what happens when you go at it alone and create your own standards while everyone else works together on a different one.

mandarinka

26th March 2013, 18:12

paradoxical

26th March 2013, 19:01

Well, there could be less problems this time around. At least since the acquisition, VP9 has been developed with the knowledge that the decoder and encoder is going to be open source and bitstream format is public. So they were probably more careful to side-step patents. Whether they were clever enough to do it properly (and whether they spotted all the hidden rocks), that is a question.

There could be but Google quite firmly stated that VP8 would have no patent issues before the MPEG-LA and Nokia came along. I think until things shake out in a couple of years that any reasonable person who doesn't buy into the PR spin would be skeptical that VP9 will not have its own set of patent issues. And that really is the danger of the going-it-alone route. At least with industry consortiums you can be reasonably confident that the players are all putting their cards on the table with respect to patents (obviously this has not been 100% true but it has worked out well for quite a few standards) since it is to each players own benefit.

JEEB

26th March 2013, 19:24

Ugh. While relatively unrelated (except for the Nokia IPR guy saying that they weren't part of that IP deal with MPEG-LA during the question time), this talk (http://recordings.conf.meetecho.com/Recordings/watch.jsp?recording=IETF86_rtcweb_II&chapter=part_4) of a seemingly Google person about VP8 for WebRTC at the IETF contains blatant lies, over-exaggeration as well as twisted facts. Gentlemen, grab your pop corn!

Oh, and it seemingly was supposed to be a technical talk.

P.S. It also contains a quote from Cisco that they would release a "good" real-time oriented AVC/H.264 implementation under a "less limited license than the GPL" in case AVC/H.264 would end up being selected by the IETF.

xooyoozoo

27th March 2013, 01:49

Some folks are trying to find prior art (http://www.groklaw.net/article.php?story=20130324162902177) to Nokia's patent claims.

While I wish them luck, I doubt they'll find much of value. Nokia's known to choose its patent battles wisely, and the company has a storied history in image/video R&D.

Ugh. While relatively unrelated (except for the Nokia IPR guy saying that they weren't part of that IP deal with MPEG-LA during the question time), this talk (http://recordings.conf.meetecho.com/Recordings/watch.jsp?recording=IETF86_rtcweb_II&chapter=part_4) of a seemingly Google person about VP8 for WebRTC at the IETF contains blatant lies, over-exaggeration as well as twisted facts. Gentlemen, grab your pop corn!

Yikes, that was a painful watch...

dapperdan

27th March 2013, 15:17

Yikes, that was a painful watch...

From the comments I was expecting discussion about H.264 vs VP8 on technical quality, but that links' mostly about IPR issues.

I wasn't really impressed with any of the argument being made at this meeting, it was all a bit political in the sense that they couldn't just come out and say "we're not going to support anything with patent royalties" and "we don't care if you're not going to support anything with patent royalties" so there was a lot of useless noise about peripheral issues like quality and hardware support in an attempt to sway the floating voters.

However, I did think the video codec comparison in this next video was, by quite some distance, the worst video codec comparison I'd ever seen which, if you've seen some of the stuff published on blogs whenever video codecs become a current news item, you'll realise takes some doing:

http://recordings.conf.meetecho.com/Recordings/watch.jsp?recording=IETF86_rtcweb_III&chapter=part_1

mandarinka

18th May 2013, 15:27

http://www.infoworld.com/d/open-source-software/googles-open-video-proposal-closes-door-software-freedom-218765

Simon Phipps (OSI) on the draft of VP8's patent license. According to him it is potentially incompatible with free software licensing, because it doesn't allow sublicensing and each receiver is required to formally enter into the agreement with Google.