4 ms·

Interesting article. Although I think it is an overreaction. Obviously MPEG-LA could be trying to enforce their patents against everyone, but they aren't. Given

by bcl 16y ago

Interesting article. Although I think it is an overreaction. Obviously MPEG-LA could be trying to enforce their patents against everyone, but they aren't. Given the level of adoption of H.264 if they started to do so they would quickly kill their potential income stream. The answer isn't for everyone to go back to the dark ages of mjpeg, it is to convince MPEG-LA to be reasonable.

nitrogen 16y ago
I dislike broadly-worded mostly-unenforceable laws and contract provisions almost more than plain bad laws and contracts, because then instead of pleasing a law, you have to worry about pleasing whoever is in charge of enforcing the law. If the enforcers don't like you, they can find something you're doing wrong and charge you with it/sue you for it. So, what's not troubling to me is not the possibility of MPEG-LA suing everybody, but rather just suing people and companies they don't like (i.e. competitors, detractors, etc.), since virtually everybody making a corporate training video with an H264-based camera is violating the camera's H264 license.
- GHFigs 16y ago
  rather just suing people and companies they don't like This is what the "non-discriminatory" part of the "reasonable and non-discriminatory" licensing terms that ISO/ITU require of their participating members is about.
  - ZeroGravitas 16y ago
    Clearly that's not true. They do sue people (they list them on their website) and yet, as discussed here, they don't sue everyone who breaks the ridiculously over-broad terms. That implies a choice. Then how do they discriminate here? I don't know, but it's not dictated by RAND terms which are often not reasonable either. Currently they also discriminate against smaller companies due to yearly caps on fees, and discriminate against certain business models e.g. subscription TV pays distribution fees, ad supported Hulu doesn't. That last one is clearly a decision made to mess with competition in a market with fewer barriers to entry while cementing lock-in in other areas.
ZeroGravitas 16y ago
If you're relying on MPEG-LA being reasonable then you'll be disappointed to hear that they are now patent trolling: http://news.ycombinator.com/item?id=1306309 http://news.ycombinator.com/item?id=1306309 That's not hyperbole, and it's not the same as chasing RAND patent fees on standards as they have previously done. They're just patent trolling the same as any other non-practicing entity. Currently suing HTC, RIM and Apple for smartphone patents. The solution, as noted already in this thread, is for governments to ban fees on standards, just as they currently don't allowy non-RAND fees on standards. China is pushing heavily for this and MPEG are at least appearing to consider it: http://news.ycombinator.com/item?id=1310388 http://news.ycombinator.com/item?id=1310388
- anamax 16y ago
  > The solution, as noted already in this thread, is for governments to ban fees on standards, just as they currently don't allowy non-RAND fees on standards. Huh? What law says that if a standards body choses a technology, the patent owner must RAND? If that were actually the case, folks would set up standards bodies to force RAND. Note that the same applies to fees, only moreso.
  - ZeroGravitas 16y ago
    Well in China, that may actually be the case, the government seem to be positioning themselves to compulsory purchase the patents if they can't be amicably licensed or worked around. But what I was proposing more moderate, that the governments only adopt "Open Standards" as they are known. If there's a patent identified then either the owner gives it up freely, or it is worked around. This is the exact analogue to current systems where those involved in the standardization agree to RAND fees or the patent is worked around. This does nothing to prevent other patent holders fouling things up, but that's true of the current system too. To make this work well you'd probably need a patent estoppel system where it is the responsibility of the patent owner to identify standards that infringe their patents rather than the other way round. If they fail to notify in time, then they'd lose the patent or at the very least the right to profit from the network effect of the standard (see this paper for details http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1134000 http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1134000). But it can work without that extra legal support and there's plenty of Open Standards to prove it.

by bcl 16y ago

nitrogen 16y ago

I dislike broadly-worded mostly-unenforceable laws and contract provisions almost more than plain bad laws and contracts, because then instead of pleasing a law, you have to worry about pleasing whoever is in charge of enforcing the law. If the enforcers don't like you, they can find something you're doing wrong and charge you with it/sue you for it. So, what's not troubling to me is not the possibility of MPEG-LA suing everybody, but rather just suing people and companies they don't like (i.e. competitors, detractors, etc.), since virtually everybody making a corporate training video with an H264-based camera is violating the camera's H264 license.

GHFigs 16y ago
rather just suing people and companies they don't like This is what the "non-discriminatory" part of the "reasonable and non-discriminatory" licensing terms that ISO/ITU require of their participating members is about.
- ZeroGravitas 16y ago
  Clearly that's not true. They do sue people (they list them on their website) and yet, as discussed here, they don't sue everyone who breaks the ridiculously over-broad terms. That implies a choice. Then how do they discriminate here? I don't know, but it's not dictated by RAND terms which are often not reasonable either. Currently they also discriminate against smaller companies due to yearly caps on fees, and discriminate against certain business models e.g. subscription TV pays distribution fees, ad supported Hulu doesn't. That last one is clearly a decision made to mess with competition in a market with fewer barriers to entry while cementing lock-in in other areas.

ZeroGravitas 16y ago

If you're relying on MPEG-LA being reasonable then you'll be disappointed to hear that they are now patent trolling: http://news.ycombinator.com/item?id=1306309 http://news.ycombinator.com/item?id=1306309 That's not hyperbole, and it's not the same as chasing RAND patent fees on standards as they have previously done. They're just patent trolling the same as any other non-practicing entity. Currently suing HTC, RIM and Apple for smartphone patents. The solution, as noted already in this thread, is for governments to ban fees on standards, just as they currently don't allowy non-RAND fees on standards. China is pushing heavily for this and MPEG are at least appearing to consider it: http://news.ycombinator.com/item?id=1310388 http://news.ycombinator.com/item?id=1310388

anamax 16y ago
> The solution, as noted already in this thread, is for governments to ban fees on standards, just as they currently don't allowy non-RAND fees on standards. Huh? What law says that if a standards body choses a technology, the patent owner must RAND? If that were actually the case, folks would set up standards bodies to force RAND. Note that the same applies to fees, only moreso.
- ZeroGravitas 16y ago
  Well in China, that may actually be the case, the government seem to be positioning themselves to compulsory purchase the patents if they can't be amicably licensed or worked around. But what I was proposing more moderate, that the governments only adopt "Open Standards" as they are known. If there's a patent identified then either the owner gives it up freely, or it is worked around. This is the exact analogue to current systems where those involved in the standardization agree to RAND fees or the patent is worked around. This does nothing to prevent other patent holders fouling things up, but that's true of the current system too. To make this work well you'd probably need a patent estoppel system where it is the responsibility of the patent owner to identify standards that infringe their patents rather than the other way round. If they fail to notify in time, then they'd lose the patent or at the very least the right to profit from the network effect of the standard (see this paper for details http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1134000 http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1134000). But it can work without that extra legal support and there's plenty of Open Standards to prove it.