> While Nokia's complaint against Amazon concerned streaming video, its complaint against HP involved the company's desktop and laptop computers, which allegedly contained the same proprietary H.264 and H.265 technology.
I have not read Nokia's brief, but that is particularly eyebrow raising.
Is Microsoft shipping h264 as a OEM MS Store app like they do for HEVC these days? Either that, or they must be shipping bloatware with AVC, otherwise it seems the buck would stop with Intel/AMD/Nvidia/Microsoft, not HP.
All this is just another nail in the coffin for VVC though. No one big enough to get sued wants to touch it with a 10 foot pole, and Nokia has justified the paranoia once again.
I thought that with patents, each entity among the manufacturing chain potentially needs a license? There is no analog for the first sale doctrine, or neither a requirement for consistent enforcement (as is there for trademarks).
Certainly not for everyone. Even if you buy professional videoconferencing equipment for business purposes, you do not get a patent license for its commercial use. Those paid offerings typically come with terms similar to the ones Cisco includes in their free Open264 builds:
THIS PRODUCT IS LICENSED UNDER THE AVC PATENT PORTFOLIO LICENSE FOR THE PERSONAL USE OF A CONSUMER OR OTHER USES IN WHICH IT DOES NOT RECEIVE REMUNERATION TO (i) ENCODE VIDEO IN COMPLIANCE WITH THE AVC STANDARD (“AVC VIDEO”) AND/OR (ii) DECODE AVC VIDEO THAT WAS ENCODED BY A CONSUMER ENGAGED IN A PERSONAL ACTIVITY AND/OR WAS OBTAINED FROM A VIDEO PROVIDER LICENSED TO PROVIDE AVC VIDEO.
I'm not sure if if the relevant patent pools even have incoming licensing coverage for live Internet video because that arrived only after the pools had long been established.
There was a fad of issuing a "covenant not to sue" in lieu of a license, specifically to try to avoid triggering exhaustion, but courts seem to be (correctly, in my IANAL opinion) treating them as licenses anyway. See, e.g., https://www.patentlyo.com/patent/2013/01/3rd-circuit-covenan...
I would not say "easily"... patent holders will absolutely try to avoid the effects of exhaustion by imposing post-sale or field-of-use restrictions, but lawsuits about it have gone to the Supreme Court, e.g., https://supreme.justia.com/cases/federal/us/553/617/ (and not worked out well for the patent holder).
Nokia must be desperate for cash.
Developing video codecs is a painful exercise as some part of the implementation almost always relies on prior art. The company that I worked for were sued for our implementation of an MPEG-2 codec, because of a similar patent owned by a troll.
These cases are what hold innovation back and are only attractive to the trolls because of how many h.264/h.265 codec instances are running out there.
The usual technique is to go for the little guys first and then move up the food chain.
> The patents were at the center of a legal dispute involving Lenovo in 2020, in which the computer maker claimed Nokia did not fully disclose its intellectual property interest to the ITU and ISO while the bodies were working to standardize H.264. That lawsuit was settled one year later, with Lenovo and Nokia agreeing on a cross-licensing agreement.
That looks like patent trolling to me: sue someone who’s not really in the business and settle so when you sue a bigger player you can tell the judge that this other company did the right thing and paid for your incredibly critical technology.
I am not a lawyer, so: even if a patent expires, that doesn't make patent violation commit in the past "not a crime anymore" does it? It's still subject to the statute of limitations?
And if they knew their patents were being violated but didn't sue, isn't there also the "sue's it or loose it" clause that can be invoked to get the case thrown out?
"Did you know, for example, that content optimization on your device’s screen when switching between portrait and landscape video is a Nokia invention? And fast forwarding or rewinding a video by scrolling through it, while simultaneously displaying the current scene"
A VCR would simply display the frames that happened to pass by the read heads while rewinding.
When doing that digitally, your codec decides if that is even feasible. If the codec optimizes for disk space, generating any frame requires going through all previous frames. If the codec optimizes for fast scrubbing, the file must contain redundant information and thus be larger.
Nokia is claiming that h264's checkpointing mechanisms allow for that (empirically, this is indeed a feature of h264) and it is therefore their invention. I'm unaware if previous codecs had similar mechanisms. It would be wierd of none had, but it wouldn't surprise me that much.
MPEG2 has I, P, and B, frames, so unless they "refreshed" the patent is somehow old, someone at Nokia just came up with the scrubbing idea using the existing implementation.
...But video players have had scrubbing bars for a long time? I'm not sure how long, my memory on Windows on XP and older is fuzzy.
> generating any frame requires going through all previous frames
No it doesn't; that would be crazy. There are keyframes that can be decoded independently of any other frame. They're usually every few seconds at most, which is fine for scrubbing.
It's not perfect, but h264 is generally "pretty good", and presumably its patents should be expiring in the next few years. At that point, we will (hopefully) finally have a codec that's supported basically everywhere, has pretty decent "quality to size" ratio, and will cost nothing to license.
Of course, by that point I'm sure everyone will have moved onto HEVC or something, but a guy can dream of this utopia.
AV1 looks very cool but it doesn't have nearly the level of compatibility that H264 has. H264 works out of the box on browsers, as well as having hardware support on phones and SBCs and set-top-boxes and whatnot.
AV1 might eventually get there, and that would be awesome if it does, but H264 has been around long enough to have a substantial "first mover advantage" over AV1.
Yeah, no question; H264 had a 10+ year head start and entrenched itself as part of the Blu-ray and online video sharing space. Once AV1 has similarly-largely-scoped things then it'll probably take a lot of h264's market, and it'll be royalty-free from the get-go.
I definitely want AV1 to become the "video standard to rule them all", but once H264's patents expire I think I'm also ok with it being the standard as well.
Last Raspberry doesnt have H264 decode hardware, wouldnt be surprised if that becomes a trend. A lot of SBCs only support up to 1080 on h264, 4k requires h265.
Yeah, I read that, though I think the latest Raspberry Pi might be fast enough to decode h264 in software so it might be a non-issue anyway; as long as it is fast enough I think I actually prefer software decoding just because (I think) it's more platform-independent.
I think I get your point though; if the Raspberry Pi is not adding chips for native H264 decoding, set top boxes and other SBCs might drop it as well.
As I have been rambling about it for years. A truly patent free Video Codec that is good enough.
I also think we could innovate on top of it. Using certain techniques form JPEG-XL, ( Royalty free and not Patent encumbered ), breaking certain limitation it had like the size of macroblock. And lots of other limitation it had in place simply because hardware at the time wont fast enough.
This could push another 30%+ Bitrate reduction. As EVC Base Profile has shown to be possible.
Or we might do it the other way round and build a video codec based on JPEG XL.
I never finished it, but during a hackathon I did try and make a codec around FLIF (which more or less merged with/morphed into JPEG XL IIRC). I didn't get too far, but it was one of those things that I didn't really understand why no one else had done it.
So I think a codec based around JPEG-XL sounds cool. Someone should do it.
Hopefully standardization committees in future will have learned to commit to open source, open licensing, mutual patent retaliation, or at least some method to prevent one defector from actually suing all the rest.
I think if you sit on your thumbs for 20 years while MPEG LA sell "your IP" without your permission, and in such a globally obvious way, you forfeit all right to that IP. Especially at the eleventh hour, like this. Especially when companies like Nokia are so used to the patent process. They have no excuse here.
Lenovo were idiots to settle. They fed the troll and now it's back.
55 comments
[ 0.24 ms ] story [ 74.2 ms ] threadhttps://www.nokia.com/blog/nokia-seeks-compensation-for-amaz...
I have not read Nokia's brief, but that is particularly eyebrow raising.
Is Microsoft shipping h264 as a OEM MS Store app like they do for HEVC these days? Either that, or they must be shipping bloatware with AVC, otherwise it seems the buck would stop with Intel/AMD/Nvidia/Microsoft, not HP.
All this is just another nail in the coffin for VVC though. No one big enough to get sued wants to touch it with a 10 foot pole, and Nokia has justified the paranoia once again.
Otherwise... thats pretty horrific. Would businesses using HP laptops en masse be exposed too?
THIS PRODUCT IS LICENSED UNDER THE AVC PATENT PORTFOLIO LICENSE FOR THE PERSONAL USE OF A CONSUMER OR OTHER USES IN WHICH IT DOES NOT RECEIVE REMUNERATION TO (i) ENCODE VIDEO IN COMPLIANCE WITH THE AVC STANDARD (“AVC VIDEO”) AND/OR (ii) DECODE AVC VIDEO THAT WAS ENCODED BY A CONSUMER ENGAGED IN A PERSONAL ACTIVITY AND/OR WAS OBTAINED FROM A VIDEO PROVIDER LICENSED TO PROVIDE AVC VIDEO.
https://www.openh264.org/BINARY_LICENSE.txt
I'm not sure if if the relevant patent pools even have incoming licensing coverage for live Internet video because that arrived only after the pools had long been established.
There is. It is called patent exhaustion. See https://en.wikipedia.org/wiki/Exhaustion_doctrine_under_U.S....
There was a fad of issuing a "covenant not to sue" in lieu of a license, specifically to try to avoid triggering exhaustion, but courts seem to be (correctly, in my IANAL opinion) treating them as licenses anyway. See, e.g., https://www.patentlyo.com/patent/2013/01/3rd-circuit-covenan...
Stating the obvious: That's hardly the case here.
> The patents were at the center of a legal dispute involving Lenovo in 2020, in which the computer maker claimed Nokia did not fully disclose its intellectual property interest to the ITU and ISO while the bodies were working to standardize H.264. That lawsuit was settled one year later, with Lenovo and Nokia agreeing on a cross-licensing agreement.
That looks like patent trolling to me: sue someone who’s not really in the business and settle so when you sue a bigger player you can tell the judge that this other company did the right thing and paid for your incredibly critical technology.
And if they knew their patents were being violated but didn't sue, isn't there also the "sue's it or loose it" clause that can be invoked to get the case thrown out?
When doing that digitally, your codec decides if that is even feasible. If the codec optimizes for disk space, generating any frame requires going through all previous frames. If the codec optimizes for fast scrubbing, the file must contain redundant information and thus be larger.
Nokia is claiming that h264's checkpointing mechanisms allow for that (empirically, this is indeed a feature of h264) and it is therefore their invention. I'm unaware if previous codecs had similar mechanisms. It would be wierd of none had, but it wouldn't surprise me that much.
...But video players have had scrubbing bars for a long time? I'm not sure how long, my memory on Windows on XP and older is fuzzy.
No it doesn't; that would be crazy. There are keyframes that can be decoded independently of any other frame. They're usually every few seconds at most, which is fine for scrubbing.
Meanwhile e.g. Apple has very strongly enforced similarly silly UX patents. (E.g. rubberband/inertial scrolling and quite a few more.)
They have also very strongly enforced silly design patents - e.g. for rounded corners on physical devices.
https://ffii.org/nokia-and-airbus-elected-as-judges-at-the-k...
This Court is a Kangaroo one.
Of course, by that point I'm sure everyone will have moved onto HEVC or something, but a guy can dream of this utopia.
AV1 might eventually get there, and that would be awesome if it does, but H264 has been around long enough to have a substantial "first mover advantage" over AV1.
I definitely want AV1 to become the "video standard to rule them all", but once H264's patents expire I think I'm also ok with it being the standard as well.
I think I get your point though; if the Raspberry Pi is not adding chips for native H264 decoding, set top boxes and other SBCs might drop it as well.
https://news.ycombinator.com/item?id=38083588#38088009
TL;DR It’s neither patent free nor ”patent safe”.
I also think we could innovate on top of it. Using certain techniques form JPEG-XL, ( Royalty free and not Patent encumbered ), breaking certain limitation it had like the size of macroblock. And lots of other limitation it had in place simply because hardware at the time wont fast enough.
This could push another 30%+ Bitrate reduction. As EVC Base Profile has shown to be possible.
Or we might do it the other way round and build a video codec based on JPEG XL.
So I think a codec based around JPEG-XL sounds cool. Someone should do it.
Then again, I think IP is a bad paradigm that does more harm than good in general
Lenovo were idiots to settle. They fed the troll and now it's back.
nokia vs hp u.s lawsuit document: https://www.courtlistener.com/docket/67928650/1/nokia-techno...