Tumbled Logic

Apr 23 2009

Paint on a canvas

We don’t know a huge amount what Project Canvas—if it gets BBC Trust Approval—will entail.

What we do know is that the BBC is keen to build on existing standards where possible (and following the failure of Kangaroo at the market impact evaluation phase, this isn’t surprising), but that doesn’t mean they don’t have an agenda.

My predictions:—

  1. Canvas will be built fundamentally on DVB-IPI. That is, DVB streams transported over IP. More than likely, wrapped up in RTP streams. How channels get multiplexed will probably depend on a whole lot of factors.

  2. To make it possible in a world of monthly usage caps, the BBC will push ISPs hard to let them co-locate concentrator nodes.

  3. Video will almost certainly be Dirac, rather than H.264. It’s worth noting that BBC R&D is pushing to have Dirac ratified as the VC-2 standard, and if they’re successful, that would very likely mean that we’d start to see it on Blu-Ray discs in a couple of years.

  4. Audio could be anything, but AAC’s a good bet.

  5. “Red Button” content will remain MHEG-5 (as it is on Freesat and Freeview) for the time being, simply from an ease-of-authoring perspective, and because STB manufacturers already know how to work with it.

  6. There will be concerns (and pressure) surrounding license-fee-payers. More on this in a separate post, though. It’s easy enough to resolve, provided people don’t revolt against the Government over it and in reality, it’s a canny move to preserve license-fee income.

  7. Stream capture will be not especially easier (and no more difficult) than with DVB-T/DVB-S, aside from hardware.

  8. Eventually, Canvas will replace iPlayer (or rather, iPlayer will get rolled into whatever fruits Canvas bears).

  9. The Canvas “platform” (outside of the technical standards) will ultimately amount to a source directory; anybody who follows the specs will be able to start broadcasting, but if you want it to be easy for people to find your channels, you’ll need to have it listed in the official (and automatically-subscribed-to) directory.

  10. The BBC is clearly frustrated with Freeview/Digital Switchover progress, and Canvas will give it the distribution platform that it’s been hankering for without the massive levels of investment required to take on Sky or Virgin on their own turf.

My take is that the BBC are well aware that they’re effectively being held hostage by Adobe, and that it’s putting very real barriers in place right now. This isn’t sustainable, and it’s not going to sit well with the BBC Trust. Flash’s terrible video playback simply isn’t a good basis for a long-term investment in an IP-based delivery platform to rival Freeview/Freesat.

I have an inkling that there are more than a few people within the BBC who view iPlayer much the same way that I do: all credit to the teams responsible for it, as the results have been impressive, but compared to what could be achieved with open and standard alternatives, Flash’s efforts look pretty poor in terms of actual video delivery.

My final prediction (for today): Ofcom isn’t going to have a clue how to regulate the result. Dealing with the BBC, ITV and Channel Four is fine, but once you take spectrum out of the loop, there are going to be real problems in defining who is and isn’t (in legal terms) a broadcaster (will YouTube be? what about the channels in Miro? you get the idea).


blog comments powered by Disqus
Page 1 of 1