First published on Biz2, 04/00.

The message of Napster, the wildly popular mp3 “sharing” software, is
plain: The internet is being turned inside out.

Napster is downloadable software that allows users to trade mp3 files
with one another. It works by constantly updating a master song list,
adding and removing songs as individual users connect and disconnect
their PCs. When someone requests a particular song, the Napster server
then initiates a direct file transfer from the user who has a copy of
the song to the user who wants one. Running against the twin tides of
the death of the PC and the rise of application service providers
(ASPs), Napster instead points the way to a networking architecture
which re-invents the PC as a hybrid client+server while relegating the
center of the internet, where all the action has been recently, to
nothing but brokering connections.

For software which is still in beta, Napster’s success is difficult to
overstate: at any given moment, Napster servers keep track of
thousands of PCs, holding hundreds of thousands of songs which
comprise terabytes of data. This is a complete violation of the
Web’s current data model — “Content at the center” — and Napster’s
success in violating it points the way to an alternative — “Content
at the edges”. The current content-at-the-center model has one
significant flaw: most internet content is created on the PCs at the
edges, but for it to become universally accessible, it must be pushed
to the center, to always-on, always-up Web servers. As anyone who has
ever spent time trying to upload material to a Web site knows, the Web
has made downloading trivially easy, but uploading is still needlessly
hard. Napster relies on three networking innovations to get around
these limitations:

• It dispenses with uploading and leaves the files on the PCs, merely
  brokering requests from one PC to another — the mp3 files do not have
  to travel through any central Napster server.
• PCs running Napster do not need a fixed internet address or a
  permanent conenction to use the service.
• It ignores the reigning Web paradigm of client and server. Napster
  makes no distinction between the two functions: if you can receive
  files from other people, they can receive files from you as well.

Leave aside for the moment the fact that virtually all of the file
transfers brokered by Napster are illegal — piracy is often an
indicator of massive untapped demand. The real import of Napster is
that it is proof-of-concept for a networking architecture which
recognizes that bandwidth to the desktop is becoming fast enough to
allow PCs to act as servers, and that PCs are becoming powerful enough
to fulfill this new role. In other words, just as the ASP space is
taking off, Napster’s success represents the revenge of the PC. By
removing the need to upload data (the single biggest bottleneck to
using the ASP model for everything), the content-at-the-edges model
points the way to a re-invention of the desktop as the center of a
user’s data, only this time the user will no longer need physical
access to the PC itself. The use of the PC as central repository and
server of user content will have profound effects on several internet
developments currently underway:

• This is the ground on on which the Windows2000 vs. Linux battle will
  be fought. As the functions of desktop and server fuse, look for
  Microsoft to aggressively push Web services which rely on content-
  at-the-edges, trying to undermine Linux’s hold on the server market.
  (Ominously for Linux, the Napster Linux client is not seen as a
  priority by Napster themselves.)
• Free hosting companies like Geocities exist because the present
  system makes it difficult for the average user to host their own web
  content. With PCs increasingly able to act as Web servers, look for
  a Napster-like service which simply points requests to individual
  users machines.
• WAP and other mobile access protocols are currently focussing on
  access to centralized commercial services, but when you are on the
  road the information you are likeliest to need is on your PC, not
  on CNN. An always-on always-accessible PC is going to be the
  ideal source of WAP-enabled information for travelling business
  people.
• The trend towards centralized personalization services on sites like
  Yahoo will find itself fighting with a trend towards making your PC
  the source of your calendar, phone book, and to do list. The Palm
  Pilot currently syncs with the PC, and it will be easier to turn the
  PC itself into a Web server than to teach the average user how to
  upload a contact database.
• Stolen mp3’s are obvious targets to be served from individiual
  machines, but they are by no means the only such content category.
  Everything from wedding pictures to home office documents to amateur
  porn (watch for a content-at-the-edges version of persiankitty) can
  be served from a PC now, and as long as the data does not require
  central management, it will be more efficient to do so.

This is not to say that desktop will replace all web servers — systems
which require steady backups or contain professionally updated content
will still continue to work best on centrally managed servers.
Nevertheless, Napster’s rise shows us that the versatility of the PC as
a hardware platform will give the millions of desktop machines currently
in use a new lease on life. This in turn means that the ASP revolution
will be not be as swift nor will the death of the PC be as total as the
current press would have us believe. The current content-at-the-center
architecture got us through the 90’s, where PCs too poorly engineered to
be servers and bandwidth was too slow and variable to open a pipe to the
desktop, but with DSL and stable operating systems in the offing, much of
the next 5 years will be shaped by the rise of content-at-the-edges.

Napster has joined the pantheon of Netscape, Hotmail, and ICQ as a software-cum-
social movement, and its growth shows no sign of abating any time soon. Needless
to say, anything this successful needs its own lawsuit to make it a full-fledged
Net phenomenon. The Recording Industry Association of America has been only too
happy to oblige, with a suit seeking up to a hundred thousand dollars per copyrighted
song exchanged (an amount that would be on the order of a trillion dollars, based on
Napster usage to date). Unfortunately for the RIAA, the history of music shows that
when technological change comes along, the defenders of the old order are powerless
to stop it.

In the twenties, the American Federation of Musicians launched a protest when The
Jazz Singer inaugurated the talkies and put silent-movie orchestras out of business.
The protest was as vigorous as it was ineffective. Once the talkies created a way to
distribute movie music without needing to hire movie musicians, there was nothing
anyone could do to hold it back, leading the way for new sorts of organizations that
embraced recorded music — organizations like the RIAA. Now that the RIAA is faced
with another innovation in distribution, it shouldn’t be wasting its time arguing that
Napster users are breaking the law. As we’ve seen with the distribution of print on
the Web, efficiency trumps legality, and RIAA needs to be developing new models that work with electronic distribution rather than against it.

In the early nineties, a service called Clarinet was launched that distributed news-
wire content over the Net, but this distribution came with a catch — users were never
ever supposed to forward the articles they read. The underlying (and fruitless) hope
behind this system was that if everyone could be made to pretend that the Net was no different from paper, then the newspaper’s “pay directly for content” model wouldn’t be challenged on-line. What sidelined this argument — and Clarinet — was that a bunch of competing businesses said, literally, “Publish and be damned,” and the Yahoos and News.coms of the world bypassed Clarinet by developing business models that encouraged copying. But other companies developed new models well after realizing that Clarinet’s approach was wrong, and they still took years to get it right. The idea that people shouldn’t forward articles to one another has collapsed so completely that it’s hard to remember when it was taken seriously. Years of dire warnings that violating the print model of copyright would lead to writers starving in the streets and render the Web a backwater of amateur content have come to naught. The quality of written material available on-line is rising every year.

The lesson for the RIAA here is that old distribution models can fail long before
anyone has any idea what the new models will look like. As with digital text, so now
with music. People have a strong preference for making unlimited perfect copies of the music they want to hear. Napster now makes it feasible to do so in just the way the Web made it possible with text. Right now, no one knows how musicians will be rewarded in the future. But the lack of immediate alternatives doesn’t change the fact that Napster is the death knell for the current music distribution system. The music industry does not need to know how musicians will be rewarded when this new system takes hold to know that musicians will be rewarded somehow. Society can’t exist without artists; it can, however, exist without A&R; departments.

The RIAA-Napster suit feels like nothing so much as the fight over the national speed
limit in the seventies and eighties. The people arguing in favor of keeping the 55-MPH limit had almost everything on their side — facts and figures, commonsense concerns about safety and fuel efficiency, even the force of federal law. The only thing they lacked was the willingness of the people to go along. As with the speed limit, Napster shows us a case where millions of people are willing to see the law, understand the law, and violate it anyway on a daily basis. The bad news for the RIAA is not that the law isn’t on their side. It plainly is. The bad news for the RIAA is that in a democracy, when the will of the people and the law diverge too strongly for too long, it is the law that changes. Thus are speed limits raised.