Okay, so the record industry and the feds are committing industrial genocide on US-based internet radio (with a few exceptions that include popular public radio stations like WUNC and KUOW). But there are plenty of places in the world where webcasting is still legal, and nobody's keeping you from listening. So let's stop for a moment and ponder the opportunities here.

On the one hand, the easiest way to put up an internet radio station is with Linux and other open-source tools. In fact, it says here that Bill Goldsmith of Radio Paradise is making his hacks available for anybody who wants them. For background on how cool this is, see "The Promise of Radio Paradise: An Open-Source Challenge to Commercial Radio".

The client side is another matter. One reason internet radio is still at the early adopter stage (or earlier: more like hunter-gatherer) is that computers are the only available receivers. Kerbango made a run at pioneering the internet radio receiver business a couple years back, but 3Com bought and killed them ($81 million for nothing). Penguin Radio came along around the same time and is still in business. Their radios look cool, but they aren't out yet.

Yet the advent of WiFi begs radios. Wouldn't it be cool to have a portable radio, or a car radio, that picked up web stations over WiFi or a wired DHCP connection to the Net? Hey, why not?

Here's what Bill Goldsmith said when I asked him about the prospects:

Then I pointed him to zradio, a piece of $10 shareware that turns your Linux-based Sharp Zaurus into an internet radio receiver. He replied, "Well, that is just the coolest thing...I could easily write a web app for that screen size that would let me control the station as well as listen to it--anywhere with WiFi access. How cool is *that*?"

But what we need now is a radio--not a computer or a PDA, simply a radio. A dedicated device. Something anybody could use.

It's silly to wait for the consumer electronics cartel to make one. Their Hollywood connections probably forbid it outright, but the bigger reason is production economics. They don't make anything unless it risks selling huge numbers.

Many years ago I did a lot of work with Hitachi Semiconductor, where I often heard the expression, "10-10-10." No, it wasn't fertilizer. It stood for an unspoken policy that was common throughout the semiconductor industry: the big consumer electronics component producers wouldn't bother to ramp up production on a new microprocessor if it wasn't going to be over 10 MIPS, under 10 dollars and ready to sell at least 10 million units. That's why the Hitachi SuperH never would have been available as a GNU/Linux development target if the company hadn't sold piles of early versions to companies like Sega and Nissan.

So we have to roll our own. Hey, is it really that hard? We don't need something beautiful, just something functional. The innards could consist of a low-power board running Linux, an analog board for audio, some backplane for USB, Ethernet and PCMCIA--or only with WiFi and nothing else.

We have to think inventively here, and cheap, too. Here's Don Marti:

You can forget venture capital in any case. Here's the RenguinRadio CEO Andrew Leyden:

That last question is a rhetorical one, because he's pressing ahead with development anyway. But what about the rest of us? Just because the going got tough doesn't mean the tough should stop going. It seems to me there's an opportunity here. What do the rest of you think?

Doc Searls is senior editor of Linux Journal.