Well,after years of lousy debates on web, the video on Web looks to be standardized on H.264 than WebM.
Here is the Mozilla CTO’s point of a view on that.
As a person who has embraced open source and open standard, I am very friendly to open source. In S.Korea, I and others in this field have thought that US was a nation of open source to exaggerate a little.
However, I found out that this country was very much polarized in this industry also.
At Harris, I found out that people didn’t like open sources. They thought it was low quality and irresponsible piece of code. I said it was different from projects to projects.
However, at current company, I found out that people blindly followed open source. After talking with them, they didn’t seem to understand what open source really is in terms of their state.
Whether things are open source or not can’t be a major critical factor to judge a project or a technology is good or bad.
There are many factors to judge that : quality of code, how well they are managed, what is the goal of the project, how easy it is to take the open source and use it in your project, how well S/W architecture is constructed, etc.
Why HTML 5 people decided to standardized on H.264 is due to the trend in the industry. One of the things I know is that H.264 is MPEG 4 AVC. In other words, it’s MPEG. There are many codecs and file wrappers for video formats used in the video industry. By saying video industry what I mean is to embrace all : Creation, Editing, Distribution, Consumption.
Where WebM and thus Theora make sense is the “Distribution” part. However, H.264 is very convenient format for video creation/editing and distribution even for consumption because most of the codecs and file formats are derivatives of MPEG and its file wrappers. Of course, we can transcode one to other. Or we can use intermediate codec or raw audio video stream and re-encode to other formats from the raw format. However, if things are based on MPEG and its derivatives, although there are still transcoding is required, for example between MPEG-2 and MPEG-4 and many different profiles of MPEG-4, the effort becomes minimal. It lessens lots of burdens on shoulder of people in the video field and saves lots of time.
People who only see “PC-based distribution” seem to care about WebM/Theora. They are good formats, too. And people can defend themselves, “There were already a lot of video codecs and file formats. Adding one doesn’t make it worse.”
True. I agree with that. However, the trend is to reduce those to make the workflow simpler. There are so many formats : XDCAM, XDAM PRO, XDCAM HD, IMX, DV 25, DV 50 blah balh, DVC Pro, DVCPRO HD, DVCPRO HD Pro, DVCPRO Progressive, DVCPRO 50, … to name the few.
They are basically based on MPEG. Strictly speaking, all video codecs are influenced by MPEG. So, once you follow what’s happening in MPEG, you can understand other codec also. However, codecs in MPEG family have their own continuity. So, it’s better to be standardized on MPEG.
So, that is why big players in the industry, which also make H/W devices, preferred H.264. If distribution channel is also standardized on H.264, the whole process becomes a lot simpler.
At least we don’t expect a TV set to receive WebM/Theora format over the air and cables, do we?