8 ms·

We have gone quite far off course but I thought I would summarise our polite disagreements which I doubt we will resolve: 1) How much effort it would be to int

We have gone quite far off course but I thought I would summarise our polite disagreements which I doubt we will resolve: 1) How much effort it would be to introduce and support Vorbis over the long term. You seem to believe it is negligible in cost while I believe it is quite achievable if they wanted to do it but VERY far from free to support additional codecs. 2) Terminology. I don't like appropriation of words and think that terminology does matter. I was clear in an early post that I understood that the AAC/AVC/MPEG4 codecs are not free (or Free). I will continue to argue against claims (even implicit) that Open Standards are not open. 3) Innovation. Although I haven't responded to the point I don't see Vorbis as a significant innovation beyond AAC so I don't see Apple as holding back innovation by not supporting it. I don't see much innovation held back by the lack of Free codecs but perhaps you do. 4) "it benefits them to sabotage the industry and to hold back innovation" This is a really strange one to me. As (3) I'm not sure about the innovation. Also the idea that they have a duty to at their own expense (1) take actions that benefit other people seems odd to me. If they were using their own proprietary codec that would be different but I just don't see the duty that they are breaching. 5) My view is that there are advantages to the limited set of codec required to support virtually all devices that may even outweigh the disadvantages of them not being Free. This is probably our core disagreement. Until there is at least a factor of two improvement in coding efficiency it is unlikely to be worth making a leap in codecs. Vorbis, theora, VP8 all came too late (at least with hardware implementations) to win this generation and it is best that an open non-proprietary winner takes all. The presence of the Free codecs (and to an even greater degree VC-1 which at the time was pitched as royalty free) helped ensure reasonable licensing terms for AAC/AVC etc. (unlike MPEG4 part2/H.263) so for almost all use cases the cost of AAC/AVC/MPEG4 is negligible in business scenarios. (Shipping a free Linux distribution or other free player/encoder software is the only case that I can see being troubled but they mostly get around it by not including the codecs while making it easy for the user to obtain them elsewhere.)

shmerl 13y ago
1) It's not even the point whether there is some cost or none. It's not the cost that prevents Apple from implementing open codecs. It's control, I already emphasized that above. They spend a lot on other things, it's not like they direly lack resources. 2) I didn't invent the usage of "open" as in "liberated". That's used in open Web, open technologies and etc. Usually some try to complain that free software is different from open source software and so on. While it's formally valid, most often you can deduce what's implied from the context. Open codecs imply liberated codecs. Open standard (in many contexts) includes being not encumbered by patents (surely, even closed codec can be implemented as open source, it doesn't make it free because of that). And that's not the case with proprietary codecs - they are patent encumbered. That's why they are not open standards. When something is discussed as open vs closed as in not having a public specification, it's usually pretty clear. 3) Vorbis is clearly innovative over MP3. We were talking about digital players, and most of them use MP3, not AAC. Opus is clearly innovative over any above, including AAC. 4) I already explained, it's not about expenses. It's about control. It's not just about "taking actions that benefit other people". It's about taking the opposite - actions which damage others (i.e. HTML5 codecs mess). Apple can't escape being blamed for it. 5) Even if there is some benefit in limiting the number of codecs, it's not what drives Apple not to use open ones, especially when they are technically better. So we agree that Apple doesn't see a benefit in using open codecs, but we disagree on their motivation or what exactly they see as a benefit. You view them equating benefit with reducing expenses. I think they consider control as the primary benefit.
- josephlord 13y ago
  (1) Even Apple prioritise and focus. Do you have evidence besides the bank balance that it isn't cost but control? (2) I'm not saying you invented the use of open or that Vorbis isn't open but that the word applies at least equally to traditional open standards. Open standard does not mean that there are now patents and the software people new to standardisation who believe otherwise are wrong. The attempt to appropriate the word (which you didn't start but you are continuing) isn't an attempt to debate on the merits. Patent encumbered standards can be compatible with many proper open source licenses. (3) I don't know Opus or how good it is but it seems too new. The proposed standard RFC is only a year old and silicon specification lead times are often several times that. All iPods support AAC in MP4 containers and have since the first iPod shipped so that is why I was comparing with AAC, MP3 is just for backwards compatibility in my view. (5) Do you have any evidence for this.
  - shmerl 13y ago
    1) Aren't recent patent events demonstrate that Apple is clearly paranoidally interested in control? Monopolistic grip = control. Banning competitive technologies = control. Codecs are just in line of Apple's usual behavior. Vorbis isn't even new. You could say that VP8 is relatively new and Opus is very new, so Apple is too slow to adopt them. But Vorbis is nowhere new. Apple always showed clear disdain towards open codecs, and it was never based on technical merits. Closed codecs equals control (or to make it more clear patents = control). 2) Arguments about what's an "open standard" aren't new either. I hold - open standards means unencumbered by patents. Especially when applied to the idea "open Web requires open standards". Luckily, W3C adopted that position. 3) Ubiquitous common denominator in audio players is still MP3. They aren't forward compatible though, i.e. players which support just MP3 can't play AAC. 5) Clear evidence is that other manufacturers successfully implement open codecs. It's not a downside, it's a benefit (for them and for their users).
    - josephlord 13y ago
      I don't think anyone else is reading and doubt that we will change each other's minds. But I'll give you one more response. 1) No. Sort of but I'm not sure they have achieved a monopoly. When have they banned competitive technologies? An alternative perspective is that they felt their technology and design were ripped off improperly and responded with the legal remedies open to them. Even so that show that their codec decisions were about control of anyone else. Vorbis is old but newer (at least the integer version) than AAC which matches it for quality. VO8 is newer and not much better than AVC. Opus is just too new. Apple doesn't use closed codecs, I won't concede the language to you. 2) W3C requires royalty free open standards which is great but does not make that the definition of open standards (although it is within the definition). 3) Yes and Apple support that too. It is old and poor by modern standards but I'm not sure of the relevance to this debate. 5) That is nice. I have the feeling that many of them may drop it in subsequent models if it was inconvenient. Again I'm not sure of the relevance of this to our disagreements.

shmerl 13y ago

1) It's not even the point whether there is some cost or none. It's not the cost that prevents Apple from implementing open codecs. It's control, I already emphasized that above. They spend a lot on other things, it's not like they direly lack resources. 2) I didn't invent the usage of "open" as in "liberated". That's used in open Web, open technologies and etc. Usually some try to complain that free software is different from open source software and so on. While it's formally valid, most often you can deduce what's implied from the context. Open codecs imply liberated codecs. Open standard (in many contexts) includes being not encumbered by patents (surely, even closed codec can be implemented as open source, it doesn't make it free because of that). And that's not the case with proprietary codecs - they are patent encumbered. That's why they are not open standards. When something is discussed as open vs closed as in not having a public specification, it's usually pretty clear. 3) Vorbis is clearly innovative over MP3. We were talking about digital players, and most of them use MP3, not AAC. Opus is clearly innovative over any above, including AAC. 4) I already explained, it's not about expenses. It's about control. It's not just about "taking actions that benefit other people". It's about taking the opposite - actions which damage others (i.e. HTML5 codecs mess). Apple can't escape being blamed for it. 5) Even if there is some benefit in limiting the number of codecs, it's not what drives Apple not to use open ones, especially when they are technically better. So we agree that Apple doesn't see a benefit in using open codecs, but we disagree on their motivation or what exactly they see as a benefit. You view them equating benefit with reducing expenses. I think they consider control as the primary benefit.

josephlord 13y ago
(1) Even Apple prioritise and focus. Do you have evidence besides the bank balance that it isn't cost but control? (2) I'm not saying you invented the use of open or that Vorbis isn't open but that the word applies at least equally to traditional open standards. Open standard does not mean that there are now patents and the software people new to standardisation who believe otherwise are wrong. The attempt to appropriate the word (which you didn't start but you are continuing) isn't an attempt to debate on the merits. Patent encumbered standards can be compatible with many proper open source licenses. (3) I don't know Opus or how good it is but it seems too new. The proposed standard RFC is only a year old and silicon specification lead times are often several times that. All iPods support AAC in MP4 containers and have since the first iPod shipped so that is why I was comparing with AAC, MP3 is just for backwards compatibility in my view. (5) Do you have any evidence for this.
- shmerl 13y ago
  1) Aren't recent patent events demonstrate that Apple is clearly paranoidally interested in control? Monopolistic grip = control. Banning competitive technologies = control. Codecs are just in line of Apple's usual behavior. Vorbis isn't even new. You could say that VP8 is relatively new and Opus is very new, so Apple is too slow to adopt them. But Vorbis is nowhere new. Apple always showed clear disdain towards open codecs, and it was never based on technical merits. Closed codecs equals control (or to make it more clear patents = control). 2) Arguments about what's an "open standard" aren't new either. I hold - open standards means unencumbered by patents. Especially when applied to the idea "open Web requires open standards". Luckily, W3C adopted that position. 3) Ubiquitous common denominator in audio players is still MP3. They aren't forward compatible though, i.e. players which support just MP3 can't play AAC. 5) Clear evidence is that other manufacturers successfully implement open codecs. It's not a downside, it's a benefit (for them and for their users).
  - josephlord 13y ago
    I don't think anyone else is reading and doubt that we will change each other's minds. But I'll give you one more response. 1) No. Sort of but I'm not sure they have achieved a monopoly. When have they banned competitive technologies? An alternative perspective is that they felt their technology and design were ripped off improperly and responded with the legal remedies open to them. Even so that show that their codec decisions were about control of anyone else. Vorbis is old but newer (at least the integer version) than AAC which matches it for quality. VO8 is newer and not much better than AVC. Opus is just too new. Apple doesn't use closed codecs, I won't concede the language to you. 2) W3C requires royalty free open standards which is great but does not make that the definition of open standards (although it is within the definition). 3) Yes and Apple support that too. It is old and poor by modern standards but I'm not sure of the relevance to this debate. 5) That is nice. I have the feeling that many of them may drop it in subsequent models if it was inconvenient. Again I'm not sure of the relevance of this to our disagreements.