TV Go Home says Odlyzko
New stuff from The Reg's favorite comms brainiac
It's always a delight to report new research from our favourite comms brainiac Andrew Odlyzko, and a fresh draft of Andrew's ruminations on the future of TV over Internet has popped up at his page at AT&T Labs, where he's of course a chief scientist.
Er, OK... Who? So what? And what am I even doing indoors on a day like this, you might ask...
Well, regular readers will know that barely a day passes without us finding an opportunity to plug Odlyzko's prophetic and much celebrated research paper Content Is Not King. It could have saved the bubble economy much grief with its simple proposition - and it's one you really knew all along - that people really want to pay for a communications infrastucture that lets them communicate with each other, rather than pay for one that serves up warmed-over 'content'. In other words, given the tools, we'd much rather create our own content, cheers all the same.
Which may seem pretty damn obvious when you compare the recent disaster of mobile phone content (WAP) and the equally unpredicted success of SMS phone text messaging, but be kind. As Odlyzko points out, it took thirty years for Bell's telephone to be seen as a device for person-to-person communications rather than dial-a-weather-forecast services.
So it looks like as a society, we're thinking faster than a hundred years ago... but perhaps not fast enough to save the next speculative technology bubble of "third generation" data telephony 3G. $157 billion has been spent in Europe alone to pump-prime this particular 'new, new' economy - and its implementers are all set to make the same mistakes as the most maladroit, shiny-eyed Internet hucksters.
TV is particularly interesting for a few reasons, since recent developments have blown away many of the assumptions on which network TV planners traditionally work. Although it's hardly talked about, the TV business is facing an incipient crisis every bit as significant as the one that Napster posed for the music industry.
With cheap affordable broadband Internet (a technology which ironically was devised as a medium for shoving video-on-demand over copper phone lines, only in a very controlled fashion), and TiVo style devices, consumers are suddenly a lot more discerning about what they choose to pay for, and when.
The challenge for broadcasters then is to find a mass market business model that serves premium content (over and above the reliable staples such as porn and sport), to a predictable audience. The fun part for us is that they don't have a clue.
What devices will people be tuning in on, and what transmission mechanisms will be viable? Odlyzko offers some clues.
"The Internet is likely to have a a much larger impact on TV than TV will have on Internet backbones," he predicts. "By the time TV moves to the Internet, data traffic will likely to be so large that streaming video will not dominate it."
That's because of a bandwidth glut, but it's also based on the assumption that people will be downloading individual clips rather than dropping into (and out of) pre-configured streamed content.
The pace of specific technology innovations can't be ignored he adds, and in this case Odlyzko reckons that cheap hard disks are particularly significant. It's the cheap fast hard disk that made TiVo possible, and the suggestion is that store-and-record devices are going to be particularly significant.
What he doesn't say, forgivably as its outside the scope of this particular paper, is whether these 'TV capture' devices will be more like computers, or more like TVs. Or maybe both. Either way, strategic planners can't place bets on either.
Odlyzko's terrific on putting this kind of stuff into perspective - with, for example, the observation that Hollywood turns in a year what the telcos see in revenue in a fortnight - so enjoy it for yourself here.
Alas, if only Ma Bell's planners followed the advice of its scientists. AT&T Wireless (now fully spun off from Phonezilla) launched its packet data GPRS service in the US a fortnight ago following exactly the suicidal pricing model that Odlyzko warns against... ®
Register Bootnote: Belated thanks to polymath David Mery for alerting us to the paper earlier this year.
Sponsored: Benefits from the lessons learned in HPC