Multiple companies, including Apple (and Nokia, which ain't exactly a minor player in the mobile market) objected to HTML5 mandating support for a particular codec, largely on the grounds that we don't really know the patent situations of any of the allegedly-unencumbered codecs.
Meanwhile, multiple people objected on the grounds that mandating a current (or, really, several-years-old now since that's what it is) codec in a spec that's not expected to go final for at least a few more years, and which has an expected useful life of around a decade, is just frankly stupid. It'd be like having a spec used today mandate XBM as the standard image format because that was the least-proprietary thing available 15 years ago when early browsers were being written.
Multiple companies, including Apple (and Nokia, which ain't exactly a minor player in the mobile market) objected to HTML5 mandating support for a particular codec, largely on the grounds that we don't really know the patent situations of any of the allegedly-unencumbered codecs.
Now it's my turn to call bullshit. "We don't really know the patent situations of $x" could be used as an argument against ANY piece of software or standard $x. Unless there is real evidence for such concerns, it's FUD.
There ain't no such thing as a free codec. At least, not as long as software patents exist.
Does Google want a Free, interoperable web? Then they should take the money they'd spend re-encoding all of YouTube into VP8 and instead spend it on lobbying to eliminate software patents. Then they could just use whatever's the best option from a technical perspective and we could stop having codec shitstorms every six months.
This is what groups like the MPEG-LA want us to think, but I'm not so sure. The Ogg Vorbis codec used for WebM audio has been in use for a decade, and has shipped in dozens of software and hardware products, some from large companies with big pockets. MPEG-LA made the same vague threats about patent pools against Vorbis, but they never followed through.
Xiph.org conducted a patent search early in the Vorbis process, and believes Vorbis does not infringe on any patents. Google has done their homework on VP8 as well. If they did it right, then VP8 is no more vulnerable to unknown patent threats than any random piece of software. (Sadly, any random piece of software is somewhat vulnerable.)
For that matter, there's no guarantee that H.264 is invulnerable from patent trolls who aren't members of the licensing pool. MPEG-LA doesn't indemnify licensees against third-party patents.
Sadly, any random piece of software is somewhat vulnerable.
Any random piece of software is vulnerable.
Look, if Google's serious about the threat software patents pose to openness, there's an obvious thing they should be doing, and it isn't "switch the video codec we use in our web browser". Until I see them doing some serious (i.e., big-money) lobbying to abolish software patents, I'm going to assume the whole openness thing is just marketing bullshit designed to play into geeks' stereotypes of them and Apple.
Not that I disagree, but a "patent search" early in the process for Vorbis is not that comforting. Vorbis has been around for a while now and new patents are awarded that all the time that are used against prior art. Unless Google/On2 has an inside man at the patent office raising Vorbis as prior art, it's likely that someone could craft a patent specifically intended to target Vorbis, get it approved, and then sue lots of people. Trolls take this approach fairly often.
Multiple companies, including Apple (and Nokia, which ain't exactly a minor player in the mobile market) objected to HTML5 mandating support for a particular codec, largely on the grounds that we don't really know the patent situations of any of the allegedly-unencumbered codecs.
Meanwhile, multiple people objected on the grounds that mandating a current (or, really, several-years-old now since that's what it is) codec in a spec that's not expected to go final for at least a few more years, and which has an expected useful life of around a decade, is just frankly stupid. It'd be like having a spec used today mandate XBM as the standard image format because that was the least-proprietary thing available 15 years ago when early browsers were being written.