• QuazarOmega@beehaw.orgOP
    link
    fedilink
    arrow-up
    3
    ·
    2 years ago

    It is both as you say, but the problem lies in its patents that are basically against development of competing codecs, so it holds back innovation in the name of keeping Google and the other collaborating big tech companies at the top.
    That doesn’t take away any of its great technical merits, it’s just that it is quite unethical

    • Ferk
      link
      fedilink
      arrow-up
      3
      ·
      edit-2
      2 years ago

      I’m not a lawyer, but after some reading, my understanding is that the reason they are patenting and then releasing a patent license (https://aomedia.org/license/patent-license/), is actually to try and avoid patent litigation.

      If a member from the AOM group (which includes Google but also several more giants) tries to litigate over one of the patents they have under the AOM patent license, then my understanding is that that member loses the rights to use the patents from the other members which are also submitted to the AOM and using the same patent license. So it’s a way to have some warranties that one of the members won’t go haywire and start using the patents to sue people over what they consider their “property”.

      But as I said, I’m not an expert on this. So do let me know if I’m wrong.

      My impression is that simply “not patenting” something is not enough to ensure that you are protected from it being patented in the future. In fact, we cannot even know how many “unknown” patents might be within any particular standard… so actually having patents on it that are backed by big players that other patent trolls might be afraid of upsetting is the only way to have some assurance that at least there’s some amount of protection. That, or waiting ~20 years for all possible patents to expire.

      In fact, it looks like even JPEG-XL is encumbered with Google patents apparently: https://github.com/ImageMagick/jpeg-xl/blob/main/PATENTS

      • QuazarOmega@beehaw.orgOP
        link
        fedilink
        arrow-up
        3
        ·
        2 years ago

        Oh gosh, I think you’re actually right, I thought that JXL didn’t have any such restrictions, but those have been put in place by none other than Google, I look like fool now don’t I? Lol

        There is still something that I fail to understand then, what is the real reason why Google scrapped JXL support from Chromium if it wasn’t to boost AVIF? To further complicate the matter, on the Wikipedia page there is mention of patent concerns only on AV1 (and consequently AVIF) and not on JXL. I was basing my arguments on the links I posted alongside the image, but I guess my research was incomplete

        • Ferk
          link
          fedilink
          arrow-up
          3
          ·
          edit-2
          2 years ago

          Sadly, with the way the patent system works, I don’t think any software is completelly free of them. Even the Linux kernel has been accused in the past of having encumbered stuff.

          It’s still possible that the reason Google scrapped JXL was to boost AVIF, since Google likely has more control over AV1 development. But I don’t think it was because of patents. Even Firefox has a lot of resistance for some reason to implement JXL… I think the problem is that AV1 being a standard for video that the browsers already support makes it very easy for them to support AVIF without too much of a dependency… and once they add it, they become more reluctant to add another new generation standard for the same thing, despite JXL being the technically superior format.

          I think the reason AV1 has a section about patents in the Wikipedia is because one of the points of that format was to try and protect it against patent litigation… and the sad truth is that there’s no 100% sure way to guarantee that any new standard is not violating some patent… According to that same article, the reason why AV1 might not be safe is not so much because of AOM’s patents but because of the patents from external unknown third parties that the standard might be infringing. That’s why there’s a risk, no matter what the AOM group does. As the wikipedia puts it, whether the standard is truly free from royalties it’s “Impossible to ascertain until the format is old enough that any patents would have expired (at least 20 years in WTO countries)”.