HDMI Forum Rejects Open-Source HDMI 2.1 Driver Support Sought By AMD

[EliasOfWaffle]

Sorry but this is not a HDMI group but a lawyer, the most problems of Foundations and Hardware groups is the lawyers that all time try to convince big chair and investors to adopt a closed model because laws allows this.
And very thanks to US supreme court.
And yes is not conspiracy, is obvious that can have Broadcom and Microsoft/Apple making shit and boycot open source project

[EliasOfWaffle]

"We need to protect your owned IP, we neeed" - Average Lawyer

[EliasOfWaffle]

Why Linux community not try to create a big petition vote about this feature and send to HDMI group?
this can work? or is a shitty innocent plan?

[EliasOfWaffle]

hdmi license x account https://x.com/HDMILicensing

[F.Ultra]

Are you sure ? My understanding was that when you check that box you are basically saying that you have independently obtained licenses for the patented materials (eg from Fluendo) and that you are *not* expecting Canonical/Ubuntu to have done so.

IOW the distro is saying "hey these are patented but not licensed by us" and you are saying "Yeah yeah whatever".

Ah, ok I might have misunderstood this then. I just remembered that they licensed MP3 and then the H.264 patent, both which fit this package. There is still the issue of violating patents by distributing patent covered software so perhaps Canonical have obtained licenses to allow them to distribute this package. I mean they are a much bigger and juicier target than us end users are.

edit: yes they did a deal with Fluendo: https://canonical.com/blog/codecs-co...-for-all-users

[F.Ultra]

Right and wrong. Patent laws have what is called "simultaneous invention" and Obviousness.

https://www.foley.com/insights/publications/2015/07/simultaneous-invention-as-secondary-evidence-of-ob/

Yes both simultaneous invention possiblity and Obviousness can be proven by clean room development.

If the patented design was not transferred into a clean room(this is by documents or people) and the people working in clean room comes up with the exact same design then the patent is based on something with obviousness so in fact invalid. This does make it very dangerous for patent holders to challenge a clean room implementation thinking that the majority of cases that this has happened the patent holder has lost their patent because the clean room work proved that the patent was obvious so not legal to patent.

https://en.wikipedia.org/wiki/S2TC

There is a issue with the clean room route around patents is that you can end up with a poorer quality replacement that not infringing. Yes this better than nothing.

Its not you cannot do clean room development to get around a patent. The result of clean room development is very different. Copyright clean room gives you implementation that you can use and leaves the original copyright holders license intact. Patent clean room you are doing the clean room to see if you can destroy the patent because you are attempting to prove the patent design is obvious so not legally allowed a patent..

Some party finds out you are doing a clean room to get around their patents be sure they will all out attack because a patent clean room is an offensive move.

Copyright and Patents both can use clean room to get around them. Patents clean room usage is for sure to piss off the patent holder because you were attempting to destroy the patent they have spent money to have so are highly likely to look at all the items they have to sue you out of existence or make your life hard. Clean room is far less aggressive of action with Copyright.

Patent law is a lot more black and white. Copyright you have the fair use greyness where you don't end up destroying the copyright works means to be sold just by clean room actions. Yes patent law also has very hard punishments as in triple to 10 times the infringement cost because you screw up your clean room setup because of willful infringement and willful aversion of the patent and willful attempt to destroy patent.

Trademarks is the part of IP where you cannot use clean room.

Patents you don't see clean room uses a lot because is very high risk/high reward thing where you mostly lose. Yes success-ed with the clean room you destroy the patent but you can also bet that party will never license you a patent again and will refuse todo business with you.

To clean room attack a patent holder you really need to make the choice that you are never going to do business with that patent holder ever again. Clean room against copyright does not have this major downside because its not offensive action.

There is a bit of confusion here. The first point of “simultaneous invention” only covers if two or more entities happen to create the same invention roughly at the same time so this does not cover cases like this where you have found out about a patent and then try to reimplement it in a clean room, that is not "simultaneous", that is after the fact. Note also that in the very case you cite that "the evidence of simultaneous invention as a whole weighs only modestly in favor of obviousness." according to the court.

Trying to falsify a patent by proving it to be obvious is viable but hard, just look at all the patent trolls that have managed to squeeze millions from the big tech companies with what all of us would deem to be below obvious patents (the classical XOR to clear the cursor is another one).

The second point with S2TC also have nothing to do with clean room. That have to do with trying to solve the same problem but in way that is different enough that it doesn't match the criteria of the patent, in this case the colour picking where S2TC picks colours at random instead of using the S3TC algorithm and since it is this algorithm that is covered by the patent, S2TC doesn't infringe. They do not create the same end result and S2TC does produce erroneous colours vs S3TC so they are not the same.

[oiaohm]

There is a bit of confusion here. The first point of “simultaneous invention” only covers if two or more entities happen to create the same invention roughly at the same time so this does not cover cases like this where you have found out about a patent and then try to reimplement it in a clean room,

It depends on how you setup you clean room. Clean room can be contracting out independent company that really does not know anything about what they are doing. Simultaneous invention roughly the same time has been stretched in different USA court cases out to 19 years apart or basically the complete distance of 20 year patent lifespan. Key point of "simultaneous invention" is that the two or more parties who invent the same thing must not know of the other parties invention in detail. Patent body without court case will resolve it if they are close in time frame with each other but if it more than 5 years apart its a court case to declare "simultaneous invention".

F.Ulra the clean room may not know they are attempting to re-implement something that patented they may have been just given a list of requirements this is where "simultaneous invention" comes into play.

Remember clean room setup. You have a dirty room and a clean room. The dirty room can know about the patented/copy-protected item. The clean room only has the censored information from the dirty room to work with. Yes clean room and dirty room can be different companies.

Trying to falsify a patent by proving it to be obvious is viable but hard, just look at all the patent trolls that have managed to squeeze millions from the big tech companies with what all of us would deem to be below obvious patents (the classical XOR to clear the cursor is another one).

Yes hard costly and painful. Clean room does come into this process as a method to attempt to prove obvious. Yes proving the "possibility simultaneous invention" by have a documented clean room party generate the same thing proves the case of obvious. It comes down to the court of the work of the clean room is declared to be just "possible simultaneous invention" or is "simultaneous invention" really the court ruling on this normal makes bugger all difference because either ruling the patent is obvious so the patent is invalid.

The worst part about this you prove that a patent is obvious you normally don't have rights to recover the costs of proving it. Yes of course patent trolls being shell companies even if the court rules you can recover costs there is normally no assets their to recover costs..

The second point with S2TC also have nothing to do with clean room. That have to do with trying to solve the same problem but in way that is different enough that it doesn't match the criteria of the patent, in this case the colour picking where S2TC picks colours at random instead of using the S3TC algorithm and since it is this algorithm that is covered by the patent, S2TC doesn't infringe. They do not create the same end result and S2TC does produce erroneous colours vs S3TC so they are not the same.

S2TC is like where clean room can land you. Think you have not given your clean room the exact details of the patent so the clean room returns you like S2TC form list of instructions based off what S3TC is in broad terms. Now you try to send new instructions to the clean room to make them return S3TC instead of S2TC you are going to find yourself in legal hell because you can end up handing over too much information to clean room so end up in willful infringement..

Clean room test with a patent not giving you back the very close to the same thing that some what works with issues can suggest that the patent is not in fact obvious. This is also what makes using clean room with patents to attempt to get around a patent a fairly much one and done. The clean room process either proves that the patent is highly obvious or it proves that its not highly obvious. Something found not to be highly obvious by clean room method still can be called obvious by a legal ruling.

Lot of the patent troll issues would be killed of both parties at the start of the patent case were required to put up bond to cover costs of the other party if they lose. Yes this would see a lot more patent cases because a party who is sure a patent is obvious or is prior art would be willing to challenge because they would know by winning they would not be out of pocket.

The biggest problem with the patent system is that is possible to go around enforcing and if someone challenges you the result is that you keep all the profits and the party who challenged loses because they cannot get the cost back of challenging. Patent system has it more cost effective to settle on patents then challenge them no matter how stupid the patents are.

[smitty3268]

Where patents apply you find another way and it leaves legal proof of that.

What makes you think it's possible to do this? I have no idea what the actual pieces involved here are, but if the HDMI spec says to do X, Y, and Z, and Z is patented, then you're kind of screwed. You can choose to do Z differently to avoid the patent, but then it's no longer HDMI and all the existing hardware that hooks up to HDMI may not work with your change.

Maybe you get lucky and you're able to make the change in a way that's compatible enough, but there's no guarantee that's possible.

It also might involve some DMCA nonsense, if it involves secret encryption keys or something similar, and DMCA explicitly disallows reverse engineering to get around that.

[billyswong]

Outside of the obvious issue that many TVs don't provide DisplayPort and force one to deal with HDMI, another issue is DP isn't really a complete replacement of HDMI.

I am not sure who added such flaw, but there are numerous reports of monitors unplugged and monitors turned-off is not differentiable in DP by design. This causes stupid PC usage annoyance that doesn't happen in HDMI/DVI/VGA.

[Anux]

The restricted package also containes DeCSS but CSS was AFAIK never patented.

That was a different case "circumventing copy protection". And it's also perfect legal to distribute the software, only if a company would use it they might get in trouble, private usage is fine (in most countries at least).

Canonical licenses H.264 - https://www.omgubuntu.co.uk/2010/05/...-count-updated I also remember that they licenses the MP3 patent before it expired but there does not seem to by any info on it left on the Internet that I could find.

In your link it says:

Neither RedHat, makers of Fedora, or Novell, makers of Suse, appear on the list of over 800 licensee’s.

At least RedHat would be as likely a target as Canonical.
And at the end:

Mark Shuttleworth has helped clarify the issue: “I believe the licenses he’s referring are purely in support of OEMs in specific cases”

Which I interpret as OEMs might have problems shipping open source HDMI but not distros itself. System76 would have problems shipping PCs with HDMI driver.

I am not sure who added such flaw, but there are numerous reports of monitors unplugged and monitors turned-off is not differentiable in DP by design. This causes stupid PC usage annoyance that doesn't happen in HDMI/DVI/VGA.

Can you be more specific? I never noticed any different behavior between HDMI and DP.