SuitWatch

September 25, 2003

The Syndication Solution

One of my ongoing writing assignments this year is to tell the story of Linux in webcasting.

So far, the big story is all on the big-time supply side. It's huge, and that's only if we take Real Networks into account. I visited Real on my last trip to Seattle (just before last week's Linux Lunacy Geek Cruise -- watch the Linux Journal Web site for reports) and I was impressed in the extreme. These guys are doing some amazing stuff with Linux -- and a lot of it. They're another example of a smart company doing the smart thing, deploying Linux back-end servers by truckload, and then by increasing the size of the market by contributing development support to the open source Helix Community.

But what about the small-time supply side? And the demand side as well? What's happening there? I've written a lot about what Bill Goldsmith has done with KPIG and Radio Paradise (see the long list of links below). But Bill is just one broadcaster. Although he's done some great hacks with Linux, MySQL and other LAMP tools, he's a smart and resourceful broadcaster, not a developer. So his hacks are still confined to just a few stations. The last time I asked Bill for an assessment of what's going on in the fledgeling small-operator webcasting industry, he wrote,

Most Internet radio is pretty heavily Windows-based, since there's lots of cheapie automation software, encoding tools, etc. available for Windows - but in the Linux world you need to "roll your own" from a variety of different bits & pieces. Someday (hopefully soon) I'd like to bring all of those pieces together into a (fairly) easy-to-use package.

He adds,

One "radio building block" in particular that you might want to feature is ecasound - a sound-processing utility that I use to do totally professional-grade dynamics processing (peak limiting/compression) on the audio.

Professional audio dynamics compression -- which is what music radio stations use to make themselves sound as loud as possible all the time -- has traditionally been expensive stuff. When I was in the radio business, back in the 70s and 80s, the big kahuna in the broadcast audio processing market was Orban. Well, Orban still holds about a 70% share of that market, and the prices of its goods run in the thousands. Meanwhile, ecasound is freeware that's GPL'd by Kai Vehmanen and currently at v2.3.0.

Still, that's just one tool, and it's just on the supply side. What about the demand side?

I've been investigating that lately, as a customer. It's an interesting perspective for me, because my first interest in the business was on the supply side, as a retailer. Many years ago, even before I worked in radio, I ran the audio department in a large department store in New Jersey. This was back when the best audio sources were vinyl and analog technology was doing its best to push distortion down to large multiples of the near-zero values that are standard in digital products today. In those days there were a pile of relatively small quality-driven companies that constantly upstaged each other setting standards for the industry. All of them competed for the attention of music lovers and audiophiles. Even when names like Scott, Fisher, Harmon-Kardon, Sherwood, Marantz and McIntosh gave up shelf space (or sold out entirely) to Pioneer, Yamaha, Sony, Sansui and Kenwood, audiophile demand had a real relationship with manufacturer supply. And it showed. Behind my desk are a Yamaha receiver and a Marantz tuner (with a scope) with FM tuners that beat anything on the market today -- not only in raw performance, but in plain usability.

Audiophiles are still around, and they still have relationships with suppliers; but now they're only driving the high-end manufacturers whose names and goods are largely invisible outside specialty magazines and audio/video salons. What's happening on the broad "consumer" side of the market is a rolling mess of disposable goods.

Here's what I found when I went hunting for gear:

  1. As a value, reception is dead. As far as the manufacturers are concened, nobody listens seriously at home to AM, or even FM, any more. And nobody gets their TV from any source other than cable or satellite. Hookups for FM antennas on the backs of receivers and table component radios are usually dumb and terrible, often putting two posts with spring tabs where a coaxial connector ought to go. The AM tuners are all awful, with terrible reception and sub-minimal audio, because the chipsets were down-costed almost to zero a decade ago. Often they come with a lousy loop antenna that the user thinks is for FM. So they hook it up to the FM terminals, where the antenna yields bad FM and no AM at all. In the town where I live, exactly none of the stores -- including Circuit City, Costco, two Radio Shacks and two high-end audio places -- had outside antennas hooked up to anything other than their satellite TV systems. And in almost every case the other display units had either no antennas, or the wrong antennas. Nobody seemed to care, because, frankly, few of the units would be used primarily for radio, which has been reduced to an automotive accessory. The main sources for home sound these days are CDs, DVDs, MP3s and cable and satellite.
  2. Thanks to the likes of Bose and Cambridge, you can get good speakers at a good price. Of course, you need to get up to six of them at a time if you want surround sound, but hey: credit where due. (Possibly relevant observation: "digital" labels aside, speakers are analog.)
  3. The high-end guys are in the Central System business. They want you to have a big thing, preferably in a rack, that goes in a closet somewhere, and drives speakers all over your house. These are then controlled by local keypads that few people can understand. We have two friends who have these systems. Neither of them understands how to use them. At one house recently we nearly destroyed the hearing of children in an outer bedroom when we turned up the volume control at the central unit, which was many rooms away.
  4. User interfaces are largely from hell. Teaching accordion would be easier than teaching how to use some of these things. And more worthwhile.
  5. There is little awareness, on the part of most manufacturers, that the Internet (much less Internet Radio) exists.

So let's go back to the supply side again. Sooner or later, whenever you talk about Internet radio, the question of business models comes up. How can a little one-person station actually make money? Where's the advertising? How many listeners will pony up through PayPal?

Wrong questions, all of them.

We actually have one business model already, in public radio. The problem is that public radio continues to operate in nearly complete ignorance of the Net and its potential for building remunerative market relationships with listeners.

This subject came up yesterday morning when Dave Winer complained eloquently about WBUR's nearly complete absence of any systems that allow real market conversations to happen between the station and the customers that keep it in business — namely, its listeners. Like most public stations, WBUR lacks a simple way for listeners to make quick & easy contributions over the Web. That's because, like most big public stations, they're obsessed with obtaining high-paying sponsorships, and subscribe to direct mail up-sell mentalities with the rest of us. One look at KQED's support page tells exactly that story.

There's an extreme irony here, because listener- and viewer- supported noncommercial radio and TV are the only forms of broadcasting that actually do business with their users. They are the only ones whose consumers are also customers. By contrast, commercial broadcasting's consumers and customers are different populations. Calling commercial broadcasting consumers a "market" is extremely misleading, because they aren't on the buying side of the marketplace. They're on the selling side, serving as the product sold to advertisers, at fees based not on what any one consumer offers, but on ratings over which only the surveyed few have any influence.

Yet, in spite of the fact that it suffers no split between its consumers and customers, public broadcasters continue to leverage the woe-is-us please-keep-us-from-dying sales pitch that Pacifica stations wore out with their arch-liberal audiences in the 1960s. It was Pacifica that broke all the ground that public radio has farmed since NPR came along in the 1970s. Pacifica pioneered listener sponsorship, the pledge drive and a host of other weary fundraising traditions, from banks of phone volunteers to parades of re-runs between breaks for appeals by station personalities and visiting celebrities.

With few exceptions, public radio and TV stations show no evidence that they even begin to understand the fund-raising potential of the Net, much less the abundant networked sources for original programming, widespread news gathering, community involvement and other virtues public broadcasters love to talk about but practice with a relative rareness that increases with every worthy form of radio that arises outside their cloistered system.

Exhibit A: Christopher Lydon's interview series. Chris is a former New York Times reporter and radio host whose show on WBUR was a favorite for thousands. Now he's doing great radio on the Web, through this interview series. Until Chris's series came along, I had no interest in carrying around a portable MP3 player. Now I want one, just so I can hear his stuff, and other stuff like it — including my own countless tapes and digital recordings of conference sessions, plus the hundred or more interviews I've done for Linux Journal and have never transcribed. Those are all sitting in boxes and on shelves right now. They belong in an archive that can be heard, on an as-wanted bases, over the webio.

Exhibit B: The RSS feed for Chris's interviews. RSS stands for Really Simple Syndication. It's the means by which everything posted on Linux Journal's web site appears in Google News, plus countless other aggregation services. Syndication is quickly becoming a fundamental Internet service -- one by which everything new and linkable can notify the rest of the Net that it now exists. Why shouldn't stations of all sizes, on both radio and TV, put out an RSS notification when a new show, or song, or anything, is up? Why not start relationships between RSS sources and the TiVos of the world? Well, only because we're leaving all the innovation up to large manufacturers who remain disconnected from what's really going on -- or worse, are trying to put a whole pile of DRM into your household entertainment systems.

Exhibit C: The Howard Dean campaign. Want to know why Howard Dean is raising so much money on the Net? Two reasons. One is that he hasn't been bought by large contributors. The more small contributions he receives, the less he owes to anybody in particular. The other is that he makes contributing easy. There's no minimum, and there are no fancy up-sells for Major Sponsors. It's not a fancy game. It's brick-simple.

There's another reason why Dean succeeds: there's plenty of involvement by volunteers. Everybody who writes for the Web is in a position to contribute to the campaign, whether they like it or not. All they have to do is operate compatibly with the natural marketplace for opinions and ideas.

That's what happened right here a few weeks ago.

On July 22, my piece called Saving the Net went up on the Linux Journal site. The next day, Matthew Gross wrote Saving the Net, and Politics, According to Doc Searls: or, Bringing out the Bat on the Howard Dean campaign weblog. The challenge: raise more money on the Web from many contributors than Dick Cheney hoped to raise that weekend at a $2000/plate fundraising dinner. The goal was to raise more than $250,000. By the time Cheney had his luncheon, the campaign had raised $344,000. By the morning of July 29, the total passed $500,000, from 9,500 people.

Disclaimer: I'm a registered independent, and completely non-partisan in my reporting on politics. But I do believe the Dean Campaign is delivering a trainload of clues about how networked markets get smarter faster than most suppliers, whether we're talking about politicians, big media, or big retailing.

So the challenge is to network the webcasting marketplace. RSS is one piece. Exposing audio archives is another. And coming up with a worthy radio is the final one. The DRMeisters of Silicon Valley and Japan aren't going to deliver it. That leaves the job up to us.

It shouldn't be hard. The source should be Ethernet and Wi-Fi. The tuner should be some kind of open directory setup, and/or something that gets informed by RSS notifications. (Example: when Terri Gross's Fresh Air comes on, I want to know.) Some kind of recording system, or interface, without DRM, is another piece. Then some knobs, speakers, amplifier and outputs to other equipment. And you've got it. Syndication and directory services are the key pieces. Work those out in an open way and we'll blow old-fashioned broadcasting out of the water.

(Oh, by the way, I am told by the Dean Campaign that they're looking for some good Linux jocks to work on staff. If you're interested, write to me and I'll pass your name along.)

Links:

Smart companies doing smart stuff

Geek Cruise

Bill Goldsmith, KPIG, Radio Paradise. ecasound, broadcast gear and Linux-based radio

Public radio, new sources, consumer electronics

Doc Searls is senior editor of Linux Journal.