Google Chrome To Remove Theora Video Codec Support

[Old Grouch]

Theora is about 2 decades old, VP8 isn't that young either. So where is the proof of those stupid patent trolls? They can claim anything they want, that doesn't make courts agree with them. So it's literally just spreading lies, panic and conspiracy theories that anything might happen. If you follow that idiotic way of thinking you can't use anything, because there can always be somebody making ridiculous claims. That doesn't mean anything.

Being loudly and vociferously convinced about something doesn't necessarily make one correct, In fact, there's an old quotation popularised by Carl Sandburg about arguing legal cases:

If the facts are against you, argue the law. If the law is against you, argue the facts. If the law and the facts are against you, pound the table and yell like hell

I hope your income stretches to assuring that you have a sufficient quantity of tables.

If software patents did not exist, we would choose codecs on technical merit.
Given that software patents do exist in some important jurisdictions, and that both 'submarine' patents exist as well as 'standard-essential' patents existing, then it is reasonable to evaluate the risk of being affected by the use of non-free technology.

Software patents, in the USA, have a lifetime of 20 years after filing.
Xiph Theora 1.0 was released in November 2008. VP* was released by On Technologies in September 2008. So both are younger than 20 years old.

Given the changes in US Patent law in the past couple of decades, I think (but I am no patent lawyer) that in December 2028 one could be reasonably certain that Theora 1.0 does not infringe any patents. until then, it's not known. If you are being cautious, you might decide to wait until all possible patents have expired before using a particular technology without a licence from putative patent holders - or have deep pockets and be prepared to fight an expensive legal case, the result of which is less important than having the means to fight it. Patent 'trolls' exist, and are usually better funded than individuals and small businesses.

[oibaf]

Software patents, in the USA, have a lifetime of 20 years after filing.
Xiph Theora 1.0 was released in November 2008. VP* was released by On Technologies in September 2008. So both are younger than 20 years old.

Given the changes in US Patent law in the past couple of decades, I think (but I am no patent lawyer) that in December 2028 one could be reasonably certain that Theora 1.0 does not infringe any patents. until then, it's not known.

libtheora 1.0 release date doesn't matter, what matter is the release date of the Theora format. And the format was declared stable since libtheora 1.0 alpha3 (source: http://lists.xiph.org/pipermail/theo...ne/001112.html), which was released in March 2004 (source: http://lists.xiph.org/pipermail/theo...ch/000950.html).

VP* is a family of codecs, and theora is based on VP3.2 (you can even losslessly convert a VP3.2 video file into a theora video file), which was itself released in August 2000 (source: https://web.archive.org/web/20071203...86&news_id=401).

So, with theora you are very safe from any patent issue.

[Artim]

Being loudly and vociferously convinced about something doesn't necessarily make one correct, In fact, there's an old quotation popularised by Carl Sandburg about arguing legal cases:



I hope your income stretches to assuring that you have a sufficient quantity of tables.

If software patents did not exist, we would choose codecs on technical merit.
Given that software patents do exist in some important jurisdictions, and that both 'submarine' patents exist as well as 'standard-essential' patents existing, then it is reasonable to evaluate the risk of being affected by the use of non-free technology.

Software patents, in the USA, have a lifetime of 20 years after filing.
Xiph Theora 1.0 was released in November 2008. VP* was released by On Technologies in September 2008. So both are younger than 20 years old.

Given the changes in US Patent law in the past couple of decades, I think (but I am no patent lawyer) that in December 2028 one could be reasonably certain that Theora 1.0 does not infringe any patents. until then, it's not known. If you are being cautious, you might decide to wait until all possible patents have expired before using a particular technology without a licence from putative patent holders - or have deep pockets and be prepared to fight an expensive legal case, the result of which is less important than having the means to fight it. Patent 'trolls' exist, and are usually better funded than individuals and small businesses.

Wow, you are just ridiculous. Go and live in the Stone Age, but don't tell others how to live their lives. Everybody will just continue using VP9 and AV1 without ever being bothered by any of the nonsense you so desperately want to see in the future.

PS: you are even too thick to provide actually true facts. Theora was first published in 2004. Sure, that was a very early alpha version, but it was already in bitstream freeze mode, so anything encoded with that version was already forward compatible. 2008 was merely the year when v1.0 was released. But saying it was released in 2008 is like saying pipewire will be released later this year or early next year, as that's when v1.0 will be released, while it has been used for years now successfully in various distributions, It's just stupid.

[Old Grouch]

Wow, you are just ridiculous. Go and live in the Stone Age, but don't tell others how to live their lives. Everybody will just continue using VP9 and AV1 without ever being bothered by any of the nonsense you so desperately want to see in the future.

PS: you are even too thick to provide actually true facts. Theora was first published in 2004. Sure, that was a very early alpha version, but it was already in bitstream freeze mode, so anything encoded with that version was already forward compatible. 2008 was merely the year when v1.0 was released. But saying it was released in 2008 is like saying pipewire will be released later this year or early next year, as that's when v1.0 will be released, while it has been used for years now successfully in various distributions, It's just stupid.

You can't patent formats.
You patent software that performs a transformative operation that is innovative.
So freezing the bitstream is not the date from which you calculate a patent's validity.
In fact, the date of publication is not the date either - (oversimplifying) it's actually from the date of filing of the patent and in the US lasts for 20 years from that date.

I'm beginning to wonder if you are a stochastic parrot, or a very small shell script.

I'm very tempted to resurrect the 'Plonk!' sound from Usenet.

[Artim]

You can't patent formats.
You patent software that performs a transformative operation that is innovative.
So freezing the bitstream is not the date from which you calculate a patent's validity.
In fact, the date of publication is not the date either - (oversimplifying) it's actually from the date of filing of the patent and in the US lasts for 20 years from that date.

I'm beginning to wonder if you are a stochastic parrot, or a very small shell script.

I'm very tempted to resurrect the 'Plonk!' sound from Usenet.

There are no patents on Theora you imbecile. That's the point of being patent free. And it's licensed under BSD, just as VP3 is that it's based on. So the only relevant thing about patent claims would have been the date the patent used for the claim was filed. And obviously these would have need to be filed before the technology was introduced in Theora/VP3, otherwise it would be invalid due to prior art. So as long as Theora didn't introduce anything fundamentally new, they would need to be older than VP3.

And as a side note, VP9 and AV1 are also both BSD 3-clause. So their release dates are also only relevant for if patents being used in claims could even be valid.

[billyswong]

You can't patent formats.
You patent software that performs a transformative operation that is innovative.
So freezing the bitstream is not the date from which you calculate a patent's validity.
In fact, the date of publication is not the date either - (oversimplifying) it's actually from the date of filing of the patent and in the US lasts for 20 years from that date.

I'm beginning to wonder if you are a stochastic parrot, or a very small shell script.

I'm very tempted to resurrect the 'Plonk!' sound from Usenet.

Your calculation may or may not be true for patents filed by the media format creator. I am not a lawyer, who knows? But if we are talking about patent troll and submarine patent, they can't be valid threat if the media format already released to public is a prior art to what they are filing for.