One of the reason that I tend to accept invitations to take part in roundtables and discussions is that I learn a lot. That was certainly the case for the session called “The Promise and Disruption of Open Source Video” that I took part in last week.
It was chaired by Michal Tsur from Kaltura. It was a company I'd already come across, but it was a good opportunity to learn a little more:
Kaltura's platform includes an easily customizable set of widgets that seamlessly integrate into video applications on web platforms of any kind. Publishers can deploy existing widgets and apps or take advantage of Kaltura's open architecture in order to develop new ones.
Whether you need a basic online video platform or one with a robust set of advanced features – Kaltura has a solution for you. Our open framework provides everything you need to design, develop, and deploy rapid and cost-effective video applications on any site.
Interestingly, Kaltura does not employ the currently-fashionable “open core” approach of providing some of its code under a free software licence, and add-ons under proprietary ones.
Instead, it's a pure dual-licensing set-up: the same code is available under free (GNU Affero GPL) and non-free licences, for all the usual reasons.
I was also familiar with another company represented on the panel, Fluendo, perhaps best known for its GNU/Linux DVD software, as well as its sponsorship of the important GStreamer multimedia framework.
Panda, though, was a startup – British, too - that was new to me:
Unlike other video platforms, Panda is not just a service for encoding your videos for the web; Panda handles the whole process. From the upload form to streaming, Panda takes control.
By providing an elegant REST API, Panda makes it completely painless to implement full video uploading, encoding and streaming functionality to your web application in a matter of hours.
As our chair rightly pointed out, open video can mean several things: open formats, open content and open tools. Perhaps the most problematic area is encodings: all the main ones are patent encumbered.
This makes the unencumbered Ogg Theora particularly important, and also means the failure of the BBC to really get behind its interesting and similarly patent-free Dirac video compression format really tragic – although maybe the open source community bears some responsibility for not doing more with it.
One consequence of this complicated multi-layered situation is that communication between the many open projects active there can be poor. But when this was pointed out in the session, it also occurred to me that one of the huge advantages of openness is that you don't really *need* to talk to other people about their projects before drawing on/working with them.
That's because (a) open licences mean you can do what you like (more or less) and (b) clean, documented interfaces at the edge of projects make it easier to link up with them. Closed solutions, by contrast, lack both of these.
So perhaps the most important thing that the open video roundtable opened my eyes to was yet another reason why openness is just so much better: it reduces the collaborative friction *between* projects as well as within them.