The means are different, the result is the same... [LWN.net]

The means are different, the result is the same...

Posted Jul 7, 2009 17:33 UTC (Tue) by khim (subscriber, #9252)
In reply to: Again: why are you so sure? by pboddie
Parent article: Ogg codecs dropped from HTML5

Even so, other companies with limited experience of pre-digital video were able to deploy digital video solutions (such as Acorn, with their Replay codecs), demonstrating that capable innovators were actually able to enter the market satisfactorily.

Yup - and where this enterprise went? Right: nowhere - to the extinction. The same people founded another enterprise (heavily-based on patents) and it's thriving today (I mean ARM Ltd).

These days, such innovators would be excluded by the kind of innuendo and veiled threats that sees open video standards excluded from open Web standards by, naturally, members of the existing patent cartels.

1. That's what I'm talking about when I talk about adoption oproblems
2. Acorn was easily excluded by other means back when Acorn Replay was relevant.

Meanwhile, well-researched (technically and legally) open codecs, such as Dirac, are the elephant in the room.

Not really. There are very few ways to produce Dirac (do you know a can which can save files in Dirac format?) while H.264 can be produced easily. There are a lot of hardware and software to play H.264 - not so with Dirac. Typical case of sunk costs. Thus I'm pretty sure dirac will not displace H.264 (like vorbis failed to displace MP3). It's good to have all these backup plans but you can be pretty sure mainstream will continue to be MP3 and H.264. And in the long run it's irreleant: by 2024 (may be earlier) all H.264-related tools will be free for anyone to use, so this particular was is lost. MP3 tools will be free by 2012 - do you still believe in future of vorbis?

The means are different, the result is the same...

Posted Jul 8, 2009 11:13 UTC (Wed) by pboddie (guest, #50784) [Link]

Yup - and where this enterprise went? Right: nowhere - to the extinction. The same people founded another enterprise (heavily-based on patents) and it's thriving today (I mean ARM Ltd).

Careful: the people who did Replay may have had a lot to do with the ARM architecture, but that doesn't mean that they wrote ARM Ltd's business model. Moreover, ARM is a hardware design licensing business, albeit with dubious aspects which probably get people into trouble even for making stuff that interprets their instruction sets. So it isn't as simple as saying that patentability makes for good business in all/any fields, even if Acorn made the mistake of not licensing their software in any sense to other people. In fact, had Acorn merely made their software available for other platforms and relied on good old copyright, it would have helped them a lot more than letting them have software patents.

Do I still believe in the future of Vorbis? Yes. The quality was better than MP3 ten years ago, at least with the tools I had available, although I really prefer lossless formats. Likewise, the myths about Theora's quality compared to the cartel formats are being put to bed as I write this. Sure, "sunk costs" ensure that there's little incentive for established companies to adopt other formats, but it's the effect on everyone else that needs addressing, at the very least.

It's easy for people to argue for more patents on the basis of giving people incentives because that's what patents are supposed to be about, but it's far from accepted that patents have been the vehicle for stimulating innovation that such people claim. Instead of going along with such claims, I suggest that the burden of proof is shifted onto those making such claims because there's plenty of evidence of the negative aspects of imposing patents on a domain.

My point about Replay undermines the assertions that people need patents to develop such stuff and that large teams in large corporations are also required. Not only are those two supposed factors independent of each other, they are also independent of whether innovation actually occurs.