WebM: Missing The Assurances Open Source Needs?
Published 07:00, 24 May 10
The announcement last week at Google IO of the creation of the WebM project and the release of the VP8 codec was a positive and welcome development, finally offering an alternative for online media to the royalty-liable H.264 and to Theora. WebM arises from Google's purchase of ON2 last year and had been widely anticipated
Google did their homework, securing endorsements from competing browser vendors Opera and Mozilla and even from Adobe (possibly in exchange for Google's endorsement of Flash on their TV platform) and, weakly, from Microsoft. The parade is now in full swing, and we can expect many more announcements of support like the one from the Miro Project. Only Apple was painfully absent, pushing the Google-Apple tension further into the spotlight
There was still more homework to do, though. Once all the hoopla had died down, it became clear there are some serious questions that need considering.
- Firstly, the new license Google is using for the project is one that's not been submitted to the Open Source Initiative for approval. As it stands it possibly can't be approved due to Google's ironic inclusion of a "field of use" restriction in the patent grant (which is restricted to "this implementation of VP8" rather than the more general grant in the Apache license from which the text started).
That means WebM is not currently open source, despite using a license based on the BSD and Apache licenses. This problem can be readily fixed by Google, and speaking as a member of the OSI Board I'd love to see them submit a templatised version of this license for approval.
- Secondly, the patent situation around WebM is unclear. Apple's Steve Jobs has previously claimed that the reason Apple does not support Ogg Theora as a video codec on the Mac is that, unlike H.264, Apple can't buy an indemnity to patent action for Theora. One H.264 partisan has already asserted there's a patent problem.
A company like Apple provides a large attack surface for a patent aggressor and their desire for insurance is understandable, even taking their antipathy for Google into account. Wanting insurance isn't an accusation of infringement. It's just a desire for mitigation of risk in a world where software patents have gone out of control.
Despite their claims that WebM was been checked for patent risks when ON2 was acquired, Google has neither made its research available nor does it offer a patent indemnity. Google has expressed extreme confidence in the patent safety of WebM, yet has failed to create a patent pool with its other endorsers and grant free and indemnified licenses to WebM users.
That means the path is open for those hostile to digital liberty, such as the MPEG-LA licensing cartel, to 'tax' VP8 users - they have already declared an intent to do so. Google should rapidly create "WebM-LA" with $0 licensing terms for those willing to commit to digital liberty.
- Thirdly, as codec consultant Rob Glidden points out, simply making the code for VP8 open source does not automatically make it an open standard. Glidden goes further, saying:
"When companies like Google ignore standards and go on their own in such important areas as video codec standards, they just undermine the very standards groups the open Web needs to thrive and grow."
We need open standards. They give us an assurance of interoperability that goes beyond the code, balancing the market power of the inventor with the freedom to leave. The 'sweet spot' of digital liberty rests at the convergence of these two 'opens', in open source implementations of open standards. Rob goes on to point out that Google could correct this issue and begin to address the patent problem in one move:
"Contributing VP8 to a standards group with a strong patent disclosure policy would be a good corrective move; it would force lurking patent holders to come fully into the public."
That of course assumes those lurking patent holders are members of the standards body in hand, but I've no doubt that a standardisation drive by Google would lead to a stronger future for WebM, not least by providing a venue for joint collaboration on liberty-compatible media standards.
Google's WebM project is a very welcome development for digital liberty. It offers the possibility for video and audio to be represented in a way that is truly open and accessible to everyone in the world, without the anachronistic licensing barriers associated with H.264.
Yet Google's do-it-all-ourselves mentality has made them forget or avoid addressing three very important issues - open source licensing, patent indemnity and open standardisation.
Over the weekend I have heard from many thoughtful people who like me want to cheer loudly yet also want these issues addressed. How about it, Google?