Visit Your Local PBS Station PBS Home PBS Home Programs A-Z TV Schedules Watch Video Donate Shop PBS Search PBS
I, Cringely - The Survival of the Nerdiest with Robert X. Cringely
Search I,Cringely:

The Pulpit
The Pulpit

<< [ The Power of X ]   |  Internet Radio is Dead. Long Live Internet Radio!  |   [ In the Eye of the Beholder ] >>

Weekly Column

Internet Radio is Dead. Long Live Internet Radio!: A U.S. Government Panel Has Recommended a Royalty System That Will Probably Kill Internet Radio and Make Possible a Better System to Follow

Status: [CLOSED]
By Robert X. Cringely

On Wednesday of this week, the U.S. Copyright Office's Copyright Arbitration Royalty Panel (CARP) delivered a report containing its recommendations on royalty terms for webcasters. The report details how much money the U.S. Government expects people who stream music over the Internet to pay to those who own the copyrights to the music being streamed. Cutting to the chase, let's just say this report means the end of Internet radio as we have come to know it in the United States. But I am not at all sure that is a bad thing.

It is a VERY bad thing in the minds of most webcasters. They complained long into the night on the mailing lists I monitor, moaning about going broke and asking each other for help with the math. And it is easy to see why these people would be so unhappy. The CARP reports recommends that webcasters pay a royalty of 0.07 cents (that's $0.0007) per song per stream for copyrighted music that is being simulcast on both the Internet and on a regular radio broadcast, and 0.14 cents (that's $0.0014) per song per stream for music that is being sent solely over the web. Non-commercial broadcasters who can't claim to be eligible for Corporation for Public Broadcasting grants, which is to say any guy who has a streaming server and just likes to stream his old records to the world for free (and also, presumably, includes all participants in systems like Gnutella, Limewire, and other peer-to-peer file sharing schemes), will have to pay 0.02 cents ($0.0002) per song per stream to simulcast music and 0.05 cents ($0.0005) for web-only songs. Noncommercial stations that DO qualify for CPB grants, which generally means all NPR stations and stations owned by 501c3 charitable or educational organizations, will continue to webcast royalty-free.

Here is why the webcasters hate these numbers. They hate them first because they exist at all; right now they are webcasting royalty-free and prefer it that way. They hate the numbers because the numbers seem excessive; it will be very hard to pay royalties AND make a profit at today's on-line ad rates. A webcaster with 1,000 streams playing 12 songs per hour for 24 hours per day would have to pay music royalties of just over $147,000 per year if all the streams were running all the time. This is more than the cost of bandwidth to carry those 1000 streams, so webcasting is going to get substantially more expensive. But they really, REALLY hate the numbers because they are retroactive to 1998, which alone will put a large percentage of webcasters instantly and permanently out of business.

Of course, this also applies retroactively to former Napster users. So if you once supplied 10,000 of your closest friends and enemies with that Roy Rogers recording of "Rudolph the Red-nosed Reindeer," don't be surprised if someone sends you a bill for $5, which doesn't sound so bad, or the minimum charge of $500 PER YEAR, which sounds VERY bad. And don't think it can't happen, either, because the very Internet archive sites that make it almost impossible for the government to remove from public access 6,600 previously unclassified documents about building weapons of mass destruction also hold records of every song transfer.

There are few loopholes, either. You can't, for example, save money by becoming the Alice's Restaurant station and playing only Arlo Guthrie's "The Alice's Restaurant Massacre," a song that is 18:20 long. CARP says royalties will be based on 12 songs per hour, so Alice's Restaurant probably qualifies as four songs.

The only real loophole is that webcasters can cut individual deals with copyright holders to get lower payments. Unknown musicians might come to such agreements to help their work become better known, but established artists often don't control their own copyrights. Major record labels might come to cut-rate terms on webcasting to get preferential treatment for their songs on the broadcast side, but that does nothing at all for pure webcasters. And as one webcaster pointed out, it is probably less effort to pay the royalties than to negotiate individually with every band.

The result of all this is a strengthening of power in the hands of larger radio broadcasters and record companies. And this should not be a surprise when it becomes clear that big broadcasters and big record companies comprised most of the content-side team that negotiated with CARP.

So webcasters this morning are preparing to switch to all-talk and all-news formats at the expense of music. They can hope that the Copyright Office rejects the CARP recommendations, which is very doubtful. Or they can hope that an alternate, discount, royalty agency suddenly appears, making it efficient for labels and artists to sign-on at lower rates, undercutting the system if it proves to be destructive. But mainly, they can prepare to kiss their servers goodbye.

Frankly, I think most of this is good. Copyright holders deserve to be compensated, and this system will make that happen. Big webcasters with ad revenue streams will simply adjust to the new reality while small webcasters probably won't survive. But the most important reason why this CARP document is good is that it will force the Internet to continue evolving, to create new methods of music distribution, rather than just emulating older technologies.

Each new medium starts out emulating the one before it, but eventually evolves into something new. Jerry Springer wouldn't work nearly as well on radio or in newspapers. Talk radio has no print analogy. So why should we expect Internet radio to sound exactly like broadcast radio, except carried over a different kind of wire?

Personally, I think the situation is ripe for the reintroduction of an XML-based version of hypertext inventor Ted Nelson's 42 year-old concept called Xanadu.

Many information technology products take crooked paths to market. That's the way it is with high technology, where great ideas usually appear years — sometimes decades — before they can become commercial products. It takes that long both to bring the cost of a high-tech product down to where it's affordable by the masses, and it can take even longer before those masses finally perceive a personal or business need for the product. Computers were around for 25 years before the price of a microprocessor dropped enough for Ed Roberts to build the first Altair computer. The BASIC programming language had been in the public domain for 12 years before Bill Gates and Paul Allen started Microsoft to do a commercial version for home computers. Fortunately, for those of us who plan to be the next Ed Roberts or Bill Gates, this means that coming up with the technical soul of our eventual empire is mainly a matter of looking up the food chain of basic research to see what's likely to be the next overnight sensation a few years from now. The technology is already there; we just have to find it.

Enter Xanadu, a system that, in its small form, is a data repository and in its large form is a publishing system. Xanadu almost made it to market back in 1992, when it was owned by Autodesk, long before the World Wide Web was even known outside of CERN.

Xanadu is a hypertext engine, a computer that is totally dedicated to storing pieces of information and links between those pieces of information, which can be in the form of text, sound, video — anything that can be represented in digital form. While anyone can have access to data on a Xanadu system, the data actually resides in only one location, and what end-users actually acquire is not the data itself, but a pointer showing them where the data resides.

This is very different from the current setup, where data is stored in computer files and the only way to share data is by copying those files again and again, filling up disk space with data that may no longer be current. "That model just doesn't work for big systems," said Nelson. "You can't reach the moon by piling up chairs."

Xanadu is different. A Xanadu document about the automobile industry might, for example, be linked directly to a database containing updated car production figures; every time another car was produced, the document would be updated. Whenever a new car model was introduced, another picture would be added to the document. Because all information is linked, rather than copied, it only has to be stored in one place and is always current. Or maybe we don't want the data to be current. Maybe we want to see the data as it was last week or last year. Well Xanadu can do that, too, acting like a data time machine, showing the way the world of information appeared to be at any given moment in the past.

One Xanadu Server is an amazing piece of software engineering, but link together two servers or a thousand or a million, and all their data and linkages can be shared transparently across the network, giving sudden access to all kinds of knowledge and insight never before available. You won't have to know where the data is stored, either, because Xanadu Server takes care of that. You only have to know what you are looking for. If this sounds a lot like The World Wide Web and search engines like Google, just imagine that Ted Nelson thought of it as a student at Harvard University in 1960!

Having linked all our servers together into a World Wide Web, there comes of problem of how to devise an information economy, how to control access to all that information and give users incentives to make their Xanadu/Servers available to others. That's Big Xanadu, the World Publishing Repository that Nelson didn't sell to Autodesk.

Big Xanadu keeps track of who owns what data, who uses what data, and acts as a giant accounting system, charging users and crediting providers for the use of their information. Publish a book on Xanadu and there are no printing or inventory costs, and royalties are paid when the book is actually read. And if my book cites data from another book or database, credit (literal credit-money) is transferred back to that source, too. "Anyone can quote anyone else, because the quote is just a link back to the original document," Nelson explained. "Not only is the original source and context of the information preserved, but the system literally pays the original author a royalty for use of his or her link in the new document."

Nelson's dream has always been that Xanadu will offer entrepreneurial opportunities to become publishers of information (put your company database on-line), to be Xanadu service providers (own one or many Xanadu/Servers on public networks), or to design and sell front-end applications to access Xanadu, giving users a viewport on all that info.

Mapping this against our music webcasting problem, it would be fairly simple to give every copyright holder a unique web site that issued XML "keys" for the playing of their music selections, the majority of whose code could live anywhere on the Internet. You could get the song from Gnutella, Napster, or on a floppy disk from your buddy, but PLAYING it would require a key that could be bought instantly from a Xanadu-type server that represented only the individual copyright holder, with the charge for playing set by the copyright holder (not by some government-sponsored blanket agreement) and carried by an efficient system like PayPal.

Once the format is agreed on, there would be no need for intermediate royalty-gathering agencies like the ones suggested in the CARP report, and no need for their overhead expenses, either. Power in the music industry would be more broadly distributed, and the people who actually own the copyrights could be better paid than they are now even if the overall royalty rates were lower.

All that is required for this to happen is to build little bits of Xanadu into the music distribution format (MP3) itself, which really wouldn't be hard. But after 42 years of waiting, Ted Nelson is probably not holding his breath.

Comments from the Tribe

Status: [CLOSED] read all comments (0)