The Gnutella paradox

As soon as an online music-trading service gets big enough to be useful, it's doomed.

Published September 29, 2000 7:18PM (EDT)

"There's always Gnutella."

If you care about music on the Net, you've probably been hearing this refrain a lot lately, repeated by MP3 traders, geek programmers and digital music industry types alike. On Monday in a courtroom in San Francisco, a judge will decide whether to uphold a preliminary injunction against Napster, potentially shutting down the music trading service -- but as a fallback, there's always Gnutella. Frightened by legal threats from the Recording Industry Association of America (RIAA), Scour may have laid off its staff and put its future in jeopardy -- but there's always Gnutella. And sure, the arcane file sharing software Freenet might not be ready for your average consumer yet -- but, of course, there's always Gnutella.

Is there, though?

Next week, Napster goes back to court to find out whether the service will be shut down for the duration of its trial. The Net is holding its breath in anticipation of the answer. If Napster is taken out, more than 30 million MP3 fans will surely flood the Net looking for a new home; Gnutella will probably be the first program many of those people download. Gnutella is not only already being heralded as the next Napster, but it's also considered by its most avid fans to be a better Napster: an open-source software program that is decentralized and anonymous, harder to sue than Napster and versatile enough to support all kinds of files.

Gene Kan, 25, Gnutella's lead evangelist and the man behind the Gnutella portal at, believes that the software is prepared for widespread use, even if he admits that it currently is still flawed. "It was really clear to us from the outset that Gnutella software had a long way to go," Kan says, but he believes that most of the program's biggest problems have been solved: "Gnutella isn't perfect, but there's no huge, glaring thing missing." And, he says, "Gnutella is very popular; it's already very successful."

But according to critics, Gnutella is hardly ready for prime time -- and is facing dilemmas almost as worrisome as the Napster lawsuit. Over the last month, users of the system have noticed a dramatic slowdown in responsiveness, and a number of reports have revealed serious instabilities in the Gnutella network. The open-source software developers who nabbed the program after America Online forced its programmers to abandon it are still striving to learn how to work together. And Gnutella's legal status is also murky: The RIAA is already hinting that it may be preparing a strategy to attack Gnutella.

Defenders of the Net love to believe that "The Man" will never be able to shut down their decentralized, "peer to peer" (P2P) way of life. Their faith is not unreasonable. File-sharing programs (not to mention chat, e-mail and other means of shooting packets of information back and forth across cyberspace) are built into the fundamental structure of the Net, and will never be entirely eradicated. But it's also quite true that corporate America can still make things very difficult for would-be challengers.

Consider this: File-sharing systems work best when they reach critical mass -- only once they have a significant number of users is it likely that someone out there will have the file you want. That's why Napster has continued to grow; with 30 million users the odds are in your favor that one or two of them will have what you need. But as soon as a file-sharing system has critical mass, it's big enough and threatening enough to become the copyright protectorate's next legal target; and those file-trading masses are also going to strain the network to its capacity and beyond. That's the Gnutella paradox. The attainment of widespread popularity may in fact signal a file trading software program's imminent demise.

If the decentralized Gnutella can't handle the legal and technical threats that come from mass usage, what system can? Or are music traders doomed to confront a future in which each new "next Napster" is progressively undermined by its own success?

It's June 1999. The programming community is shocked. Justin Frankel, the talented young programmer who helped create the Winamp and Shoutcast MP3 players, had sold his company Nullsoft to America Online. For at least a year, Winamp had been the most popular software program in the MP3 underground, one of the first tools that made it really easy to listen to music nabbed off the Net. Frankel was an icon for script kiddies everywhere, and had a history of doing whatever he felt like doing -- but selling out to AOL? Even though the price tag was rumored to be $100 million (and Nullsoft was also seeking relief from a troubling lawsuit that alleged that Winamp stole its code), many found this hard to swallow; even more suspected that AOL might not know exactly what it had gotten itself into.

For nine months, Frankel and his team worked in silence behind the corporate wall of AOL, in the company's San Francisco music headquarters. And then, one day in mid-March, the statement: a little program called Gnutella, hidden on a back page of Nullsoft's Web site. It was an early "alpha" version of what was to be an open-source (the code would be freely available to all) file-sharing system, like the increasingly controversial Napster program, but lacking the vulnerabilities -- centralized servers, lack of anonymity -- that made Napster so easy to attack.

What was Frankel thinking? AOL was in the process of merging with Time Warner, which in turn owns the EMI and Warner Music record labels. And EMI and Warner Music, as two of the five biggest members of the RIAA, are not fond of programs that allow users to pirate MP3 files. The program appeared on the Nullsoft Web site for just a few hours before AOL yanked the page down, issuing a terse statement declaring that "the Gnutella software was an unauthorized freelance project." Was Frankel trying to peeve his new corporate owners?

Nullsoft engineers had been watching the controversy surrounding Napster, and threw together Gnutella in the space of a few days as a way to prove that a decentralized system could out-geek the law. Their goal was less to annoy their new owners than to figure out how to improve upon Napster. As one person close to the Nullsoft staff explains, "They have 'fuck you money,' they can do whatever the hell they want and AOL can't take back what they gave them. I don't think that Gnutella was just done to [thumb their noses] -- AOL is insignificant. It was just the most interesting thing you could possibly be doing, AOL or no AOL."

AOL's punishment for its rogue programmers was minor: The company publicly disassociated itself from Gnutella, forbade Frankel to work on the program and hoped the embarrassment would end there. (Although Frankel, six months later, unleashed a second surprise for AOL: a little program called AIMazing, which helps eradicate ads from AOL's instant messaging program ... but that's another story.)

AOL's actions did not mean, of course, the end of Gnutella. Avid developers were savvy enough to download Gnutella before it disappeared, and before long they had reverse-engineered the program and distributed the protocols; in a matter of weeks, the Web was peppered with sites offering both the original Gnutella program and a number of clones. Six months later, more than two dozen versions of the software have been released by assorted developers.

The initial Gnutella software was hard to use: It had a confusing interface, and to connect to the network users had to scramble to find the Internet address of another Gnutella host (not always an easy task). But new versions such as Gnotella incorporated friendly Napster-like interfaces, let users design their own skins and smoothed out some basic networking issues. Shaun Sidwall, the Canadian programmer behind Gnotella, plans to incorporate a built-in host in the next version of his software, so that newbies can automatically connect to the network.

Dozens of programmers were thrilled to get a chance to tinker with Gnutella. But any technology needs its figurehead, and with Frankel hidden away in the back rooms of AOL, Gnutella needed a new spokesperson. It found one in Kan.

Gnutella -- and, for that matter, the entire P2P movement -- couldn't ask for a better representative. Like Frankel, founder Kan is a quiet and youthful programmer with a love for technology. Unlike Frankel, however, he's a master at industry diplomacy. He's young and soft-spoken and chooses his words as carefully as a law professor, excising any "ums" or "likes." He sits stiffly, with his hands in his lap, and other than his collection of zippy cars (including an RX7 and a BMW) is utterly lacking in ostentation.

Kan has done an excellent job as an evangelist: He's appeared in the pages of the New York Times debating industry heavyweights like RIAA president Hilary Rosen and antitrust attorney David Boies. He's flown to Washington to discuss policy with Sen. Orrin Hatch, R-Utah, and earlier this month attended a P2P summit organized by computer book publisher Tim O'Reilly. Thanks to all the free publicity, Gnutella's traffic has steadily grown: A recent study measured 35,000 users in a 24-hour period. Much of this growth came during the days after the RIAA won a preliminary injunction against Napster, as fans rushed to find a new program to use. (An appeals court later stayed the injunction until next week's hearings.) Kan estimates that roughly a million copies of the program were downloaded from his site that day. Today, on an average day, tens of thousands of users use Gnutella to exchange MP3 files, plus porn, pirated software "warez," illegal movies and other digital detritus, both pirated and legitimate.

But all the traffic has put a strain on Gnutella, and the program's weaknesses are starting to show. Kan, ever the upbeat evangelist for the technology, cheerfully admits that Gnutella has had its faults; but he also believes that Gnutella is ready for widespread use. "At first you focus on building the car, and once the car is built then you focus on refining the car," he enthuses. "We knew the refining was around the corner and it just takes some time. We wanted to accelerate the best we could by coordinating developer efforts and encouraging them to raise the bar on usability. And it happened."

Kan's optimism, however, is not matched by researchers who have released critical reports in the last month. A recent study by Eytan Adar and Bernardo Huberman of Xerox PARC examined the traffic on Gnutella and discovered that there were an awful lot of "free riders," users who were happy to download files but weren't willing to share their own. In fact, 70 percent of all Gnutella users don't share files, and of the 30 percent who do, the top 1 percent share 40 percent of all files. During a 24-hour period, Adar and Huberman observed 31,395 hosts -- but of that group, 314 hosts were serving the majority of the files.

It's the tragedy of the commons, writ for the digital age: Shared resources are being gobbled by users who get more than they give.

"There's very little reward for you sharing your files, and there's a high cost," explains Adar. "You're anonymous so you can't get credited for doing what you're doing -- no one says 'good job' for doing this. And it's a high cost, because [due to bandwidth limitations] even if you're on a DSL connection you can't do other things that you want to do. People realize that they come out on the negative side and don't want to share files."

The most obvious drawback of free riding is that if only a tiny percentage of users are sharing, it will be harder to find the files you want. Kan shrugs: "The providers of goods are always fewer than the consumers of goods; even if the study were true it's not such a terrible thing. Even if all the Starbucks closed, you'd still find a morning cup of coffee. Even if those people who share files are shut down, others would take their place. Gnutella didn't get here because nobody shared."

But free riding is more problematic than Kan will admit. Free riding also means that Gnutella won't scale: Each search query plugged into Gnutella has a certain "time to live," and will expire after it has queried a certain number of hosts. If the network continues to grow, and no one is providing any files, your query will hit its expiration date before it arrives at a useful host. The legal implications are also worrisome; although it's widely believed that the decentralized nature of Gnutella will make prosecuting users impossible, it's much easier to target scapegoats if only a few hundred hosts are actively dispensing files.

Adar hasn't written off Gnutella. He believes a more successful and scalable system is possible, but would require trade-offs such as decreased privacy and a few centralized network machines. Future versions of Gnutella could also include a system "default" that forces all users to share, much like Napster. Otherwise, Adar believes, Gnutella won't be able to hold up under the strain if it is flooded by Napster-like traffic.

Earlier this month, researchers from Clip2 Distributed Search Services published a report documenting another major flaw. According to the report, the Gnutella network is only as strong as the bandwidth of its weakest users -- too many people using 56K modems are being required to channel traffic, causing the entire system to slow to a crawl. The report pegs the "scalability barrier" at 10 queries per second; any more, and the network will break down.

It's not surprising that Gnutella has deep problems in its infrastructure -- after all, it was "barely in alpha" when released by Frankel. In fact, Frankel's own release notes reveal that Gnutella was designed for use by roughly only 350 "nodes" (or hosts); in a widely distributed e-mail attachment he posited that Gnutella wouldn't be scalable past 5,000 users. It's incredible that the software has survived the wear and tear of hundreds of thousands of users to this point. But it's also looking increasingly doubtful that it will be able to handle much more traffic; the system is already slowing to a crawl, and Gnutella's more devoted developers are worried about what is to come.

Eighteen-year-old Gnutella developer Sebastian Lambla of Paris is concerned. "Gnutella is not really adapted to a large number of hosts, and is not at all adapted [to] a context with spammers and flooders on the network," worries Lambla. "Free riders and slow links are two of the biggest problems of the current protocol."

Lambla has some suggestions for improving Gnutella, and he and other developers are attempting to act upon them. But because Gnutella is not owned by any single entity, the process is moving slowly -- especially compared to the increasing speed with which the recording industry is finally beginning to respond to threats. Gnutella's developers are not a cohesive group. They are scattered all across the globe, and range from teenage hackers to experienced programmers. But some are trying to form virtual networks to work on next-generation protocols and software development. Still, many programmers are prickly about letting anyone "take over" Gnutella; and when enterprising developers have attempted to spearhead the development of the client and protocol the community's reaction has been decidedly mixed.

"The general agreement is that the future designs should be by committee; there should be agreement. It doesn't help to have the babble that's initiated by independent development; it doesn't help anyone to have a whole bunch of separate systems that don't communicate with each other," Kan says. "It's a challenge, however, to coordinate the efforts of everybody; particularly because there are very, very few commercial enterprises that are devoted to Gnutella, and so it's impossible to get everyone into the same room so that they can have interaction."

As a result, there are initiatives such as gPulp -- spearheaded by Lambla -- which hopes to draw up next-generation protocols for Gnutella that will expand the system's distributed search system beyond simple file-sharing capabilities. But the gPulp group, in turn, is competing with Intel's P2P Working Group, which is also trying to develop standard protocols that will work for all P2P software products. And then there's the Gnutelladev Working Group, set up to provide services for developers who want to create their own clients.

"It's difficult to make any arrangement to get together," says Shaun Sidwall, the Canadian programmer who assembled the Gnotella client. "There's definitely some competitiveness, though; everyone would like to have the best clone out there, or the No. 1 client. Everyone likes to be No. 1, of course."

This is not unfamiliar territory -- the open-source community has been successfully tackling the issue of cooperation for years, and Linux and the Apache Web server are both excellent examples of how virtual programmers can work together toward a common end. Gnutella may well fix its technical problems in the coming months; but if Napster is shut down next week, such help won't have come fast enough.

Even if all the technical problems were fixed immediately, however, that still wouldn't put Gnutella in the clear. The enthusiastic response that first greeted Gnutella had as much to do with its seeming immunity from legal repercussions as it did to its technical, open-source backend. Gnutella's more high-profile developers would seem to be the most vulnerable to lawsuits, but even they feel relatively safe from the record industry.

Kan, for example, believes that he's safe from any blame for Gnutella-facilitated piracy: "Insofar as my role with Gnutella is concerned, I would question what benefit there would be in coming after someone like me," he says. "First off, there are many like me. And secondly, what would they get? There would be little potential for recovering any financial damage, there is absolutely zero potential of shutting down Gnutella and in any case Gnutella is nothing but a communications protocol. It'd be like suing English."

Or, as Sidwall believes, "The RIAA would have taken action by now, I bet, if they were going to."

Gnutella fans may be lowering their guard a bit too soon. The RIAA has made it clear that it does have its eye on Gnutella. Cary Sherman, general counsel of the RIAA, explains that "Gnutella is certainly one of the issues on our radar screen." Sherman says that he is already thinking about ways that Gnutella could be legally threatened, if necessary. As he observes, "Gnutella is the name of a program that is a P2P network; to the extent that there's no central source running it like Napster, it's true that an injunction can't bring it down. But there are also people disseminating the program, and people who are using it to disseminate materials. There could be legal strategies to address that." He also observes that the free-riders research done by Adar and Huberman proved just how few hosts there really were: "There's an enforcement strategy there if we wanted to pursue it," he says. "We'll monitor the situation and proceed accordingly."

The RIAA may try to make an example of Kan, Sidwall or any of a multitude of Gnutella users or developers, even if they can't shut the P2P protocol down. And if enough users are scared off, Gnutella will lose the critical mass it needs to be successful. Those who are left could potentially splinter Gnutella into dozens or even hundreds of secret sub-networks in order to evade legal scrutiny -- creating smaller groups of hosts linked together, perhaps around specific interests, rather than the one mega-network that currently exists -- but this would basically turn Gnutella into an insiders-only club: hardly the kind of mass phenomenon that would be a threat, or useful, to anybody.

The RIAA could also, conceivably, sue AOL, since AOL programmers originally created the program. But odds are that they won't. Time Warner's EMI and Warner Music are heavyweight members of the RIAA so AOL's ties to the RIAA are close, to say the least. Indeed, had Gnutella been created by anyone but AOL, many observers think that the RIAA already would be serving up legal papers.

The RIAA's Sherman scoffs at the notion that AOL is being given an easy ride by the record industry. "That's absolutely not the case," he says. But others plan to take AOL to task for its role in the creation of Gnutella., a Santa Cruz, Calif., start-up that offers several Web interfaces for finding MP3s -- including Gnutella -- is currently being sued by the RIAA. In a third-party complaint filed in early September, MP3Board struck back at AOL, claiming that if MP3Board is found guilty of copyright infringement by the RIAA, then AOL should also be found guilty since it produced Gnutella in the first place.

MP3Board's lawyer Ira Rothken argues: "AOL owns the intellectual property [of Gnutella], they own the copyright, they put it up on the Web site, they should have known it didn't have safeguards in place for Digital Millennium Copyright Act compliance. They had the right, given the fact that they own the copyright to the code, to go out and get a restraining order against people who did use it. And they didn't." So if MP3Board goes down, he thinks it's only fair that AOL goes down too -- if, that is, MP3Board can afford to battle the monolith of AOL in court.

In any case, Gnutella's fate is undeniably tied to that of Napster. If Napster wins its fight against the preliminary injunction next week, Gnutella developers will be able to breathe easy for a while; as long as Napster is still around, Gnutella developers should be able to work on fixing the software's problems without an onslaught of MP3 trader traffic. But if Napster loses to the RIAA, it will be an entirely different story.

Gnutella and Scour Exchange are the only two file-sharing applications that have even a tiny amount of the critical mass required to make the systems work smoothly. But Scour Exchange is owned by a corporate entity and is already facing a lawsuit from the RIAA; if Napster is closed then Scour will probably face the same future. Gnutella, if it survives the onslaught of millions of new users and grows into the biggest MP3 trading community, would be the RIAA's next target -- and this time, the RIAA would have a legal precedent in its pocket that declares that technologies that let users infringe on copyrights are illegal.

Yes, there will always be another P2P file exchange program -- CuteMX, Napigator, OpenNap, MojoNation and Freenet are some lesser-known programs that would happily step into Gnutella's place -- but if Gnutella, which has some of the best open-source programmers on the Net behind it, can't survive the technical or legal challenges of critical mass, how will the other programs be any better prepared?

At least that's how it looks right now. But perhaps Gnutella's problems will be fixed, and future versions will be both technically sound and truly anonymous, making it impossible to figure out who's doing what where. It's worth remembering that Gnutella is as much an ideology as it is a software program. There's a great deal of emotion invested in Gnutella, and in the long run that passion may well prevail. The subtext of the chant that "there's always Gnutella" is a giant middle finger waved in the face of a music industry that is perceived as greedy and exploitive.

As Justin Frankel himself put it in June, "Nullsoft is and was about all these good things that ultimately don't matter to most businesses. The people, the environment, the blatant disregard for conventional thinking. We did shit because it was cool, and because it was what we wanted to do." Or, as Rob Lord, one of the original Nullsoft employees, is quoted on the Nullsoft home page: "We didn't get into this 'space' cuz we're Internet gold-seeking cockos. We're legitimate nihilistic media terrorists, as history will no doubt canonize us."

Justin Frankel, the patron saint of P2P? Let the church of Gnutella endure forever. Amen.

By Janelle Brown

Janelle Brown is a contributing writer for Salon.

MORE FROM Janelle Brown

Related Topics ------------------------------------------