7 ms·

The next frontier in IP parasites: codec royalties on content

39 points by jlpcsl 3y ago

fxtentacle 3y ago
While I strongly dislike software patents, I believe they might be the only way left how a single person inventor can profit off his/her invention. Without patent protection, there's nothing stopping large companies from just using your algorithms without ever paying you, the inventor. I mean they do it to each other all the time because in many instances, racing ahead first and then settling lawsuits later is just more profitable than delaying your market entry until you negotiated all licencing agreements. So yes, we should abolish software patents. But we do need a replacement or it'll incur a massive power shift from employees to companies.
- sokoloff 3y ago
  Do you mean individuals rather than employees in the last sentence? As an employee, I already grant whatever patent rights my work for my employer creates (and I’m pretty sure that’s overwhelmingly common). But an individual inventor (not an employee) is in a different situation.
- erinnh 3y ago
  How does it shift the power from employees to companies? I get your argument for non employed individuals. But employees basically never retain their rights to the code they’ve written while employed.
- croes 3y ago
  Software patents will not help you, big companies have more time, money and lawyers than a single developer. On the contrary, your invention is more likely to be blocked because of a BS software patent such as the 1-click patent.
- chrismorgan 3y ago
  They can’t use your algorithms unless you disclose them, or they’re straightforward to invent or reverse-engineer. The patent system is (supposed to be) a trade: you tell everyone how to do something you invented, and in return your are granted a monopoly on exploiting it for a while (whether by using or selling it yourself, or by licensing same to others), after which anyone can use it. The alternative is trade secrets: you don’t tell anyone how to do it, so no one else can do it. And this works just fine with software.
- hakfoo 3y ago
  We need a stronger system to buy out patents-- not just software patents-- for the sake of retiring them in the public interest. The risk isn't just "we want 10 cents per consumption unit", it's that they have the ability to say "we won't license it for manufacturer ABC or purpose XYZ", and then they can change from being a tollbooth to progress to an actual roafblock. There should be an easy way to blow the whistle-- invite a state mediator who can say "We've calculated a reasonable value of the patent as $x.y million, take an industry-wide collection and come up with that, and the patent is voided."
xrd 3y ago
This article has a quote saying even AV1 is not safe from patent trolls. I'm interested in hearing more, I thought AV1 was as safe as it gets.
- croes 3y ago
  As safe as possible but not safe. https://news.ycombinator.com/item?id=33638975 https://news.ycombinator.com/item?id=33638975
- ksec 3y ago
  > I thought AV1 was as safe as it gets. The only safe video codec are those 30 years old codec where all patents has expired.
signaru 3y ago
I was reading about compression algorithms recently and this reminded me of the mess of LZW compression being used in GIF images.
adamzochowski 3y ago
> The first really good video codec was MPEG-4 H.264. I remember in 2001 my housemate watching a movie on his telly — playing off a CD-R. A whole movie crammed onto a CD, encoded with DivX! H264 didn't exist in 2001. At the time DivX was really a hacked Microsofts codec to allow encodes outside of .ASF files. This MS codec was a decent codec using standard Mpeg4 part 2 Simple Profile. XviD was opensource effort to ditch ms hacked codec, and eventually it went to support other profiles, like ASP (advanced simple profile). Much later some enterprising people figured that they could ride on the DivX name popularity. So they wrote own mpeg4 part2 codec and called themselves DivX LLC. Even much later this company also released h264 and h265 codecs. Hence maybe the confusion by the article author on what was used in 2001.
- ksec 3y ago
  I was about the write similar comment. People keep mixing up H.264 and MPEG-4 being the same thing ( As they did in another HN thread ). Divx in 2001 was using MPEG-4 Part 2 or better known as H.263. Divx also has its H.264 profile that came much later. Anyway I am very pissed that Nokia could be allowed to do that to AVC. And then there is next bit; >The patent trolls were the people claiming that MP3 was “dead” in 2017 — when in fact it was finally freed, because the last patents on it had expired. MP3 is sorta mediocre, but it’s also a common baseline everyone knows how to handle. And, y’know, it’ll do. The common baseline should be AAC-LC. It is also patent free as MP3, still being used in iTunes. You get much better quality while having practically same hardware and software support.
  - davidgerard 3y ago
    perhaps it should be AAC-LC, but it is MP3.
- davidgerard 3y ago
  [author here] Darnit, you're right. Correction added. I guess it must have been H.263 and who knows what for sound, probably MPEG-2. But given the video was probably 320 lines at most, it should still fit on a CD if only at "wow this is possible" quality.

39 points by jlpcsl 3y ago

While I strongly dislike software patents, I believe they might be the only way left how a single person inventor can profit off his/her invention. Without patent protection, there's nothing stopping large companies from just using your algorithms without ever paying you, the inventor. I mean they do it to each other all the time because in many instances, racing ahead first and then settling lawsuits later is just more profitable than delaying your market entry until you negotiated all licencing agreements. So yes, we should abolish software patents. But we do need a replacement or it'll incur a massive power shift from employees to companies.

sokoloff 3y ago
Do you mean individuals rather than employees in the last sentence? As an employee, I already grant whatever patent rights my work for my employer creates (and I’m pretty sure that’s overwhelmingly common). But an individual inventor (not an employee) is in a different situation.
erinnh 3y ago
How does it shift the power from employees to companies? I get your argument for non employed individuals. But employees basically never retain their rights to the code they’ve written while employed.
croes 3y ago
Software patents will not help you, big companies have more time, money and lawyers than a single developer. On the contrary, your invention is more likely to be blocked because of a BS software patent such as the 1-click patent.
chrismorgan 3y ago
They can’t use your algorithms unless you disclose them, or they’re straightforward to invent or reverse-engineer. The patent system is (supposed to be) a trade: you tell everyone how to do something you invented, and in return your are granted a monopoly on exploiting it for a while (whether by using or selling it yourself, or by licensing same to others), after which anyone can use it. The alternative is trade secrets: you don’t tell anyone how to do it, so no one else can do it. And this works just fine with software.
hakfoo 3y ago
We need a stronger system to buy out patents-- not just software patents-- for the sake of retiring them in the public interest. The risk isn't just "we want 10 cents per consumption unit", it's that they have the ability to say "we won't license it for manufacturer ABC or purpose XYZ", and then they can change from being a tollbooth to progress to an actual roafblock. There should be an easy way to blow the whistle-- invite a state mediator who can say "We've calculated a reasonable value of the patent as $x.y million, take an industry-wide collection and come up with that, and the patent is voided."

xrd 3y ago

This article has a quote saying even AV1 is not safe from patent trolls. I'm interested in hearing more, I thought AV1 was as safe as it gets.

croes 3y ago
As safe as possible but not safe. https://news.ycombinator.com/item?id=33638975 https://news.ycombinator.com/item?id=33638975
ksec 3y ago
> I thought AV1 was as safe as it gets. The only safe video codec are those 30 years old codec where all patents has expired.

signaru 3y ago

I was reading about compression algorithms recently and this reminded me of the mess of LZW compression being used in GIF images.

adamzochowski 3y ago

> The first really good video codec was MPEG-4 H.264. I remember in 2001 my housemate watching a movie on his telly — playing off a CD-R. A whole movie crammed onto a CD, encoded with DivX! H264 didn't exist in 2001. At the time DivX was really a hacked Microsofts codec to allow encodes outside of .ASF files. This MS codec was a decent codec using standard Mpeg4 part 2 Simple Profile. XviD was opensource effort to ditch ms hacked codec, and eventually it went to support other profiles, like ASP (advanced simple profile). Much later some enterprising people figured that they could ride on the DivX name popularity. So they wrote own mpeg4 part2 codec and called themselves DivX LLC. Even much later this company also released h264 and h265 codecs. Hence maybe the confusion by the article author on what was used in 2001.

ksec 3y ago
I was about the write similar comment. People keep mixing up H.264 and MPEG-4 being the same thing ( As they did in another HN thread ). Divx in 2001 was using MPEG-4 Part 2 or better known as H.263. Divx also has its H.264 profile that came much later. Anyway I am very pissed that Nokia could be allowed to do that to AVC. And then there is next bit; >The patent trolls were the people claiming that MP3 was “dead” in 2017 — when in fact it was finally freed, because the last patents on it had expired. MP3 is sorta mediocre, but it’s also a common baseline everyone knows how to handle. And, y’know, it’ll do. The common baseline should be AAC-LC. It is also patent free as MP3, still being used in iTunes. You get much better quality while having practically same hardware and software support.
- davidgerard 3y ago
  perhaps it should be AAC-LC, but it is MP3.
davidgerard 3y ago
[author here] Darnit, you're right. Correction added. I guess it must have been H.263 and who knows what for sound, probably MPEG-2. But given the video was probably 320 lines at most, it should still fit on a CD if only at "wow this is possible" quality.