Geeky notes on low-cost streaming video, linked here for my own future reference.
It’s curious that streaming video to a desktop computer, despite tiny, grainy frames, compression artefacts, badly-synced video and all the other baggage, is a wholly miraculous viewing experience. Now compare broadcast television, which does the same thing only with much better pictures (yes, despite PAL and NTSC being horrible – SECAM too if you happen to be French), minimal latency, multi-channel sound and a rock-solid frame rate, wirelessly.
Every now and then the idea of doing simultaneous net streams crops up at work, in connection with some mooted project or other. Every time, I find myself asking the producer/development researcher/exec/marketing bod/technology guru/whoever why we, a television company, would want to pay hundreds to thousands of pounds to allow dozens to hundreds of viewers to watch our output. We measure audiences in tens of thousands, not hundreds!
Not that I think streaming media is a bad thing. Far from it, I think it’s a terrific tool, and I greatly admire the sort of thing Clive James is doing with it (though that’s off-air at the moment, sadly). Also, there are several projects on the horizon where I can see myself rigging streaming into the format. A year ago, giving a presenter or actor a handheld pad-type prop which showed live streaming video would have been a nasty post-production effects job. Now, it’s quite cheap. In fact, the only link in the chain I don’t already own is the TabletPC, which – if they continue to sell so badly they’re dumped as clearance stock – wouldn’t be expensive.
No, streaming media really is cool, both as tech and as a concept. The problem is, TV has been doing it for decades already, and another transmission system doesn’t much help us at this point. There are two implications: for now, streaming is a fascinating little niche, and I expect to see all manner of clever projects taking advantage of it, but I don’t expect them to be commercial projects. Secondly, when bandwidth costs fall by – oh, say two orders of magnitude – the above analysis becomes entirely void.
We’ve already seen this with non-streaming video: early QuickTime was miraculous but utter rubbish, we all said as much. Ten years on, we’re emailing lashed-together iMovies of our loved ones.