Linux Today: Linux News On Internet Time.

InternetWorld: Who Gnu? - What Gnutella Reveals About the Flaws of Constraint Models

Jul 09, 2000, 13:20 (7 Talkback[s])
(Other stories by Nate Zelnick)

By Nate Zelnick, InternetWorld

The majority of mass-media mergers are driven by big companies' collective need to ensure their future success. Similarly, they all dislike unexpected changes in the competitive landscape that might turn the advantages that big companies generally enjoy--like established business methods, long-term relationships, and instant brand recognition--into weaknesses.

For example, the music industry shrank over the last 30 years from a few big companies and hundreds of independent labels to a handful of huge companies and virtually no independents. "Hitmen," Frederic Dannen's book on the rise and fall of CBS Records, tells how CBS, Time Warner, Sony, and their ilk boosted independent promotion--a payola system for getting music on radio stations--until it shut the small labels out. In the end, the independents were forced to sell out to the five conglomerates that now control 90 percent of worldwide distribution.

Music is the most extreme example, but the same thing is happening across the information industries. Vertical integration of content production and distribution has been the trend for more than 20 years. While those involved publicly cite illusory "synergies," the real driver is fear of change.

But gigantism is a losing battle. The technology shift of the past five years is among the most dramatic ever seen--and the impact of cheap processing power, cheap memory, cheap storage, and universal connectivity is only just beginning to be felt. These changes, taken together, mean it's only a matter of time before the landscape changes sufficiently to turn a mega media company's advantages into liabilities.

And it's time for that to happen. It only takes a quick glance to show how poorly gigantism serves both ends of the supply chain. A typical recording contract amounts to a form of modern-day sharecropping, under which the label dictates how the monies it advances an artist for promotion and production must be allocated to its collection of "company store" partners. The royalty it returns to the artist in exchange for acquiring all copyright to the music is a tiny percentage of the wholesale price the company pays itself to distribute the work. More than 90 percent of artists end up in debt to the label at the end of the contract.

Replace music with any other content and the story is a variation on a theme.

On the consumer side, the logic of bigness has resulted in the mega-hit mentality. When a label or a publishing imprint's responsibility is to contribute profits to a huge multinational, it quickly realizes that one big hit is easier to manage than lots of little successes. This creates an innate conservatism that puts all resources behind a few creators with proven selling power. This rush to the middle, of course, leaves out the building of new and interesting content.

Since entertainment depends on novelty, of course, this is also a losing strategy, resulting in frenzied premiums paid for the few dark-horse hits--or technologies that might be destabilizing--that get out through other means.

That's why the ironies of the Gnutella story are so delicious. Created by the folks at Nullsoft after AOL swallowed the company for its dark-horse hit WinAmp, Gnutella represents the antithesis of everything giants like AOL and Time Warner--which AOL agreed to buy shortly after it bought Nullsoft--stand for. Not surprisingly, once the Gnutella news reached AOL's upper echelons, Nullsoft's unauthorized project was stomped. But since AOL only found out about it after an alpha version was posted--along with the source code--it was too late. Ten thousand copies of the alpha version got out, and the Gnutella Portal Project took over further development. Thousands of servers are running all the time, and variants, ports, and extensions pop up daily.

Gnutella, Napster, and their many cousins are massively distributed peer-to-peer search-and-retrieval systems that provide no means of central control. Because there's no one in charge of Gnutella, there's no one to sue, and therefore it is even more threatening to media conglomerates than Napster, which will likely die at the hands of the Recording Industry Association of America. I love Gnutella's spirit and its technology. But Gnutella is entirely reactionary. In the end, it and similar open file-sharing systems actually will hurt artists, as the RIAA says (although it's debatable whether it will hurt them more than the recording industry itself).

What's needed is a hybrid system that gets around the bottlenecks of big media and provides a system in which creators can get paid for their work. In fact, I've resigned my position at Internet World to pursue just such an open source project. I'll still be writing this column and hope to report at some point on my progress.

Related Stories: