The virtue and vice of audio

I'm all in favor of podcasting. I think it's interesting, important and even cool (although having me be an arbiter of cool is like handing Donald Duck the wine menu). So why is it that I don't ever listen to podcasts?

Just in case you slept through the podcast flurry, let me remind you of what it is. You create an audio file and post it on your blog site, which creates an RSS summary file for it, or you post it at one of the podcast aggregators (e.g., www.ipodder.com). Someone picks up the RSS feed in her aggregator, decides she wants to listen to it, and tells her aggregator to go get it. The aggregator downloads it so that she can listen to it on her computer. Or she downloads it directly into her iPod-like device so she can listen while she jogs or drives.

This matters more than it may seem. Initially, it only means that it's easier to make audio bloggers a part of your life, which will encourage more amateurs to talk into the microphone attached to their computer. But RSS is the great leveler. As more broadcast stations make their content available as RSS feeds, it'll be as easy to add Aunt Marci's latest audio letter from home to your daily download as Reginald Oxbridge's BBC discourse on the fate of the blue-nosed bottle fly. You'll compile your playlist, put them on your iPoddy device, and listen to your personalized channel during drive-time.

So, if podcasting is so great, why don't I listen to any podcasts? It's certainly not a matter of principle: I read lots of blogs and believe having too many blogs available is the best thing to happen since radios got dials.

The main problem is that audio takes time. You can't skim audio. A summary can tell you if you think you'd get something out of listening, but there's no faster way to get that something. Peer review and reputation systems can help with this, which implies that our audio aggregators should let us rate podcasts, and should publish the ratings.

The second problem is that I work at home and thus don't have a drive time. I generally only listen to the radio when I'm in the kitchen, and that's ad hoc; I don't think about what podcasts to bring with me as I go to do the dishes.

The third problem is that I don't have an iPod-like device. My laptop is my music server, and moving it requires unplugging wires, etc. Too much work for a lazy slug like me.

So, I don't listen to podcasts. And that's a shame, not only because I'm missing some good stuff but because audio is a special medium: It's impossible to skim. When we listen, we have to hear every word in the order intended. We are taken from a start to the finish. And, of course, the nuance of the spoken word adds incredibly rich metadata to the mere content. There's a reason that "listening" implies appropriating what's being said in a way that "reading" doesn't.

This is a problem with many—nay, most—KM systems. They're devoted to the art of skimming. It starts at the system level: There is soooo much information in an organization that we should be able to skim it and present the cream. Then we should present the cream itself in skimmable form: I can check my KM portal and check the day's developments in mere seconds. If I see a headline that looks interesting, I can open it up and then skim the contents.

Skimming is obviously hugely important. But sometimes you should swim, not just skate on the surface. Audio forces you into the water. The fact that we can't skim it might just be its secret advantage.

David Weinberger edits "The Journal of the Hyperlinked Organization", e-mail self@evident.com.