3 ms·
They've dealt with one of the criticisms which was only libvpx was given a patent licence. The other major criticism was that they are still in control of VP8,
by kierank 15y ago
They've dealt with one of the criticisms which was only libvpx was given a patent licence. The other major criticism was that they are still in control of VP8, unlike H.264 which was developed and is in control by a multitude of different organisations.
- kkowalczyk 15y agoiPad was also developed and controlled by one organization, and yet it's a smashing success. Argument by correlations are weak. In your case, the argument is "standards controlled by a single organization are worse than those created by multiple organizations hence WebM is worse than h264". It's a weak argument because you can compare WebM and h264 directly on any quality you want, you don't have to use a proxy of how it was developed. Not to mention that the premise ("standards controlled by a single organization are worse than those created by multiple organizations") is easily shown to be false: there are plenty of bad standards developed by multiple parties and plenty of good standards developed by small groups. That's the origin of the "design by committee " pejorative. h264 is good not because it was developed by many companies but rather despite it.
- kierank 15y agoI didn't say anything about how the quality of a standard was affected by whether it was developed by committee or not. Google claims that VP8 is "open", yet they are in full control of it. This is unlike H.264 which is not controlled by any single person. libvpx is the main reference for the standard that is VP8 including bugs. An open standard would have allowed these bugs to be corrected before launch.