About 'casting

It has been forty-three years to the day (as I write this, yesterday as you read it) since John F. Kennedy was assassinated. Everybody alive and old enough to remember that event can say what they were doing when they found out.

I was in class at my boarding school, Concordia Collegiate Institute, in Bronxville, New York. I was a junior, sixteen years old.

At the first break between classes we went to our radios. There was no television at the school. And for the next few days there was no radio, either. Not radio as any of us knew it. Between news reports, stations played funereal classical music. Even the three Top 40 stations -- WMCA, WINS and WABC -- maintained a State of Decorum that was disorienting for teenagers like me (and most others), who lived off a constant diet of pop music.

Imagine if all iPods were suddenly filled with your choice of unhappy music from three or fewer sources. That's what it was like. Only worse, because the nation really was in a state of mourning.

This morning at my home in Santa Barbara, California, I was reminded of the Kennedy assassination by a Canadian radio station that happened to be streaming through our (Linux-based) Sonos home audio system. Earlier I had been listening to the BBC's Five Live. After both, I tuned in the Brian Lehrer show on WNYC-AM.

Most mornings I listen to the radio while making coffee and breakfast. Sometimes I listen to podcasts, but that's more complicated, so I tend to stick with radio.

The choice of stations is large, but not as large as it could be if the stations all published their stream URLs and stuck with them. After I'm done with breakfast, my wife usually puts on a classical station... WCPE from Raleigh, KUAT from Tucson, WNYC's HD Radio stream, KUSC or KCSN from Los Angeles, or one of several classical streams from the Netherlands over Avro.nl/radio. (In the old house it was usually jazz, but for some reason Classical makes better background music in the new house.) All of those station streams either come defaulted on Sonos' Internet Radio station list, or were added there by me.

We have a pretty sizeable music collection, mostly duped off CDs. If I ever get the time and energy, I'll dupe off the vinyl collection too. But there's still a funny thing about one's own collection: you tend to know all the music. No surprises. Also no connection to experts outside yourself, which is what the best of radio provides. So I listen more to radio than podcasts. At least when I'm at home. When I'm driving it's a mixture of podcasts and Sirius Satellite Radio (which is okay but has something of a canned quality to it, probably because the whole "band" is one company's silo).

My friend Howard Greenstein, an online radio veteran (he tried to make something work way too early in the dot-com era, and was less lucky than Mark Cuban), recently pointed in his blog to SpotDJ. Here's how he describes it:

SpotDJ is a plugin to iTunes that plays YOUR music, and mixes in 'spots' or small audio clips from people who know about the music and whose small clips add value to your experience. You know, like radio used to do, in the old days when DJs knew the bands they played and were able to play what they thought you'd like.

Additionally, if you know something about the bands you love and the songs you listen to, you can cut your own spots. People vote on the value of the spots so stupidity and self-promotion are voted out of the rotation.

Good idea. Problem is, it's in the iTunes silo. But as a project it still makes a couple points: 1) Anybody today can "program" anything for anybody who's interested; and 2) There's still friction involved. In the old days, "tuning" was easy. You spun a dial or hit a button. While the spinning selector works well enough with the iPod and the Sonos, it's still just a way to navigate menus, and menus are a slog.

Old-time radio still has the better UI model. Too bad it doesn't leverage. It's just a model.

And that's the thing. We're caught between a known model we're slowly abandoning and a new model we haven't worked out yet. I don't think we'll finally abandon the former until we work out the latter -- even though the former is highly expensive and increasingly absurd in an internetworked world.

Take WNYC, for example. On AM it radiates on 820KHz, from a site in the tidelands alongside the New Jersey Turnpike that was built in the 1940s for WMCA, right after WMCA won a fight for the frequency (570KHz) it shared with WNYC, which moved up the dial to 810 and 830 before it landed on 820 in the 80s, and moved to the WMCA site, where the two stations share three 335-foot towers. WMCA is 5000 watts full-time, with a signal pattern that favors New York City while protecting other stations on the same channel in Syracuse, Youngstown, Dallas, Washington, D.C. and Asheville, NC. WNYC is 10000 watts by day and 1000 watts at night, when it protects WBAP, one of the original "clear channel" stations, in Fort Worth, Texas.

Both stations spend a lot of money pumping out signals that carry only a few dozen miles before becoming unlistenable. They are typical of the whole over-the-air broadcast framework, which is based on assumptions of extreme scarcity -- of channels, of signal coverage, of real estate for setting up antenna farms.

On FM the situation isn't much better. WNYC-FM, on 93.7MHz, was knocked off the air on 9/11 when its transmitter on the World Trade Center's north tower went down (along with nearly all of the city's TV stations). The signal now is 6000 watts, radiating from the Empire State Building. The station is strong enough over the metro, but fades quickly beyond that. Unlike other cities (such as Los Angeles), New York has no large "grandfathered" huge FM signals. And even those are mostly advantaged in the nearby coverage area, where bigger signals pound through walls and into parking garages. FM signals don't carry well beyond the transmitting antenna's horizon.

VHF-TV (channels 2-13) has propagation properties similar to FM (channels 2-6 are just below the FM band). UHF-TV, with its much shorter wavelength, is much worse. While the maximum power for channels 2-6 is 100,000 watts, and the maximum for channels 7-13 is 316,000 watts. UHF signals on channels 14 and up need 5,000,000 watts to achieve "equivalent" coverage that is far from it. The new digital TV signals are down in the few-hundred-thousand watt range. In 2009 all the VHF channels will go away, and the spectrum sold off. This was actually supposed to happen in 2006, but things were moving too slowly. I have a feeling that 2009 might be just about the right time. Channels hardly matter anymore.

Spectrum shouldn't either. Back in 1927, when the Federal Radio Commission was created to regulate spectrum usage, we could hardly conceive of any other way to organize radio signals. Now we can, but we don't, because it would screw up too many big businesses, including the federal government's own monopoly at selling off spectrum.

The Net will inevitably undermine all of this mess, and will come to support a sane and useful approach to wireless connectivity. But in the meantime we're stuck with the old mentality, the old Regulatorium, and the old enforced scarcities.

The mentality may be the hardest thing to change. On cable and satellite TV, it makes less and less sense every day to simultaneously pump hundreds of live data streams that aren't being watched, and compressing the crap out of all of them to make the whole kluge work.

To move off the old broadcast framework, we need an invention that mothers necessity -- something that moves as far past the iPod as the iPod moved past the transistor radio, and that invites at least as much demand.

Ideas?

-- Doc Searls is Senior Editor of Linux Journal, a Visiting Scholar with the Center for Information Technology and Society at UC Santa Barbara, and a Fellow with the Berkman Center for Internet and Society at Harvard University.