After many months, it is with a degree of trepidation that I direct you towards the podcast audio of the ETech paper on Reinventing Radio that Mr Biddulph, Mr Hammond, Mr Webb and I performed earlier this year. As usual, listening to yourself talk is a bit of a painful experience – and it’s never a good thing to be re-exposed to jokes you made up on the spot in front of a hundred of your peers. But there you go. You live and learn. You can download the slides as a PDF here: Reinventing Radio to accompany the whole experience. Here’s a bit of a summary of the paper from the IT Conversations guys:
Isn’t radio an old, dying medium? What’s it doing in a conference on emerging technologies? Matt Biddulph, Paul Hammond, Tom Coates, and Matt Webb show us how radio is a reemerging technology experiencing a resurgence in popularity and relevance. They explore how radio can be improved by introducing feedback mechanisms and by ultimately making it a more social medium. Using principles of social software, the BBC becomes more of a peer than a broadcaster.
In the first part of the presentation Matt Biddulph and Paul Hammond explain BBC Radio’s experiment with a format called the “Ten-Hour Takeover” in which control of the station’s playlist is given over to the listeners. How can DJs be empowered with direct access to an audience of millions? With an audience that huge, how can feedback on the order of hundreds of thousands of SMS messages be handled in a meaninful way by a DJ? There isn’t enough human bandwith available to deal with that level of engagement. Traditional models would be forced to either ‘smoosh’ out the input into an average or to select a few random individuals to represent the audience. But this isn’t good enough for Matt Biddulph and Paul Hammond, who show us how they integrate SMS technology with some statistical techniques to create an ‘information space’ standing between the public and the DJ.
So you’ve got a broadcast network and you’ve got a web presence, each with very different models of interaction. How can these two models coexist in a useful and meaningful way? In the second part of the presentation Tom Coates and Matt Webb show us how radio can be enhanced using techniques from social software like flickr and del.icio.us to create a hybrid of the broadcast and network models. They wonder why we treat network computers as dumb receivers for broadcast content when they could be much more social and allow for interaction with both broadcasters and other listeners. ‘Phonetags’ bring folksonomies to radio, allowing listeners to tag songs with a cellphone as they listen. They also explore how techniques as simple as group listening can add to the social experience of radio.
5 replies on “The podcast of 'Reinventing Radio'…”
I haven’t listened to the podcast (no speakers on my work machine) but it certainly looks like a really interesting talk. “Phone tag” is an extremely neat idea. My only gripe is the phone element: 10p a shot to tag a song? What’s wrong with email, apart from it being boringly easy to compose and unfashionably cheap?
OK, I’m a late adopter wrt this mobile/pervasive stuff – I remember my then editor telling me he’d despaired when, midway through the research for a 3000-word feature on ‘m-computing’, I mailed him about these amazing phones I’d just discovered where you opened out the case lengthways and there was this like proper keyboard… (He’d had one. In his previous job. Five years ago.) I hardly ever use SMS, and when I do I use mixed case and punctuation. I also remember ’77 and have more than one LP by Hatfield and the North. I am, in short, an old fart – and as such pretty typical of the 6 Music audience, I would have thought.
How depressing that these people think radio needs improving and that the way to do it is to ‘let the audience decide’.
How long would John Peel have lasted in this world?
Radio isn’t dying and its resurgence hasn’t relied on asking the audience what to play. Slots that use playlists sent in by enthusiasts work precicely because we are hearing an alternative individual choice not some averaged out top 100 tunes.
What a terrible thought – ‘Imagine’ – played all day long!
How depressing that these people think radio needs improving and that the way to do it is to ‘let the audience decide’. How long would John Peel have lasted in this world? … Slots that use playlists sent in by enthusiasts work precicely because we are hearing an alternative individual choice not some averaged out top 100 tunes.
Not having listened to the podcast (still) I can’t be sure, but I think you’ve described precisely what the ‘infocloud’ approach is trying to avoid. One of the great things about Peel was his ability to find out what people were raving about outside the closed loop of A&R/marketing/sales. This could be a nice fat channel for carrying just that kind of information to people who aren’t quite as savvy, as dedicated or as experienced as Peel was.
But yes, by the look of it the technology could be abused. Pietro Speroni had some interesting thoughts about the (in)vulnerability of tag clouds to supplier abuse, i.e. spamming – here’s a roundup (I love del.icio.us). Perhaps we should also be thinking about user-driven abuse, i.e. cherry-picking and strip-mining. Not to mention the potential for feedback loops – give the people nothing but Keane and Dido, and an awful lot of people are going to phone in and say “Great programme, but how about more Dido and more Keane?”
I think you may have misinterepreted a bit – if you listen to the talk itself, one of our key concerns is to content that actually makes things more interesting by being able to spot tracks that people are strongly divided upon and explore that a bit. Let’s put it this way – surely no one is saying that knowing less about your audience and trying actively to not play things that they might like is a good idea? We’re just trying to find better ways of getting the audience involved in the production of radio. We specifically stated that grey programming that smooshes out all the variety from the broadcasts is a bad thing, just as randomly choosing new tracks because a few people directly requested them is a bad thing…
Funny! A comment speaking about me, but with a link to the wrong page. Oh well, the entry Phil is refering to is tag clouds are hard to spam written some 6 months ago. And when I have the time I should add an extra entry to the sequence.