Stream #0

- Sync on FFmpeg svn r21845.

Before going any further, there's a couple of things to note here. The first thing is that Lightworks is a "professional" video editing solution, in the same way that Avid Nitris and Final Cut Pro are high-end professional video editing applications. This is not iMovie or Windows Movie Maker open sourcing their code base. As such, there's probably not a lot of competition with other open source video editors, such as OneShot, PiTiVi and Kdenlive, that target the home user. Lightworks is likely to fall into the same category as Blender, Cinelerra and the yet to be released Lumiera, with high end appeal.

As I missed the previous announcement, I thought it would be a good idea to follow up and find out what's actually available from EditShare as open source at this time. And the answer is, nothing. Although the initial announcement was made in April, there doesn't yet appear to be any download available. An update from the developers in May, confirmed that their aim is to support Linux and OSx, as well as Windows. There's an interest registration page available, where presumably those who have entered their details will be informed first of updates. The page counter tells us that a little over 12,000 people have done just this. Sadly, the Beta testing programme is now closed, which was limited to just 80 people (not really the open source community way!).

Overall, exciting news and as Q3 starts next week, which is when EditShare hope to release the first publicly available open source version of Lightworks, we can all hope that it won't be too much longer before we can all play with it.

If you're interested in seeing something that's been edited using the current closed source version of Lightworks, try to find the movie Centurion.

CNET posted a reasonably ordinary piece regarding the quality of WebM, when compared against H.264. However, there were two interesting links in this piece. 

The first pointed to a WebM project page where the indepth encoding parameters for WebM content are outlined. If you're planning to create WebM files, reading this page is essential. 

The second link, to the quAVlive website provides some various examples of H.264 (using x264) encoding compared against WebM. I can't really see a lot of visual difference in the "Sunflower" example. However, it is easily clear to my eyes, without even enlarging the screenshots, that in "Park Joy" and "Riverbed" H.264 is certainly superior. I would like to have seen more information regarding the time taken to transcode these examples, with each codec, and the resulting file sizes. Picture quality isn't always everything, transcode time and storage requirements should also be taken into consideration.

Everyone's jumping on the WebM bandwagon with software and hardware support. Gstreamer claims full plugin support, which means in turn there is Moovida support and the Transmaggedon transcoder can also output VP8 codec files, although not in the Matroska/WebM container yet. Not to be outdone, Flumotion, will also stream live VP8/WebM content. The Miro Video Converter will also output valid VP8/WebM files, claiming to be the first to do so. The list could go on, but the easiest thing is to probably just keep tabs on the WebM project page listing all the supported devices and software tools, both commercial and open source.

Also worth a shout is the fact that both Mozilla and Opera are pushing for VP8/WebM to be specifically included in the HTML5 specification. Previously, major browser makers couldn't agree on one specific video file format - Mozilla and Opera backing Ogg Theora and Apple sticking with H.264. I can't see that particular situation changing now. 

Late last week this rumour became reality and WebM was born. Alongside Theora and Dirac, WebM now enters the open source HTML 5 ready codec battle. Almost immediately all major web browsers, except one, but including Internet Explorer announced support for the codec. Using the might and muscle of Google WebM must have a solid chance of taking on the dominance of H.264 in the web video delivery battle. This really will be a solid kick in the pants for Theora, which now seems destined to remain a reasonably niche product, even with direct HTML 5 support from Firefox.

In short order some early comparisons between H.264 and WebM appeared online. Some with more technical detail than others. The debate also began as to whether Google was benevolent or evil. Did WebM contain submarine patents that not even Google were aware of?

Producing WebM video for the masses was the next step. Easy to follow FFmpeg tutorials are available and just a few days ago a major commercial transcoding software vendor announced WebM/VP8 support.

WebM video is already available on YouTube, in experimental form. How long before at least all new YouTube video is transcoded to this format? If WebM quality is on parity with H.264, and the jury is still out on that, what is the unique selling point of H.264? Why would anyone continue to use it? 

There will be a substantial legacy component to overcome. Many people and organisations have invested heavily in H.264 technology, and a move to WebM may represent an operational, although not licensing, cost. However, with Google behind it, many of Big Business' concerns around open source projects may be alleviated.

Adding to this, H.264 video within a Flash player still has significant advantages over HTML 5 delivered video content, in terms of presentation flexibility and perceived security.

H.264 video is of course still dominant for web delivery, just as VP6 and VP7 was in the past. However, WebM is an exciting development with a bright future. Using the collective power of open source development, and no small amount of corporate backing from Google, watch out for WebM to challenge MPEG-LA's codec in the future.