• long Read

The Infinite Lives Of BitTorrent

It’s a way of downloading movies, it’s a startup, and with a hidden stream of revenue, BitTorrent is also a grand Silicon Valley experiment: How many times can a company reinvent itself in the hunt for a hit product?

The Infinite Lives Of BitTorrent
BitTorrent creator Bram Cohen and BitTorrent's VP of Marketing, Matt Mason, have different visions of success for the company. [Photos by Mathew Scott]

Bram Cohen sits at a round desk, surrounded by a pod of open cubicles. He looks lonely and isolated out here, hardly befitting the tech celebrity he’s become.

This is the guy who single-handedly changed the way files are transferred on the Internet. His technology helped create one of the greatest headaches in Hollywood history. He birthed the term–BitTorrent–synonymous with piracy, or a radical rethinking of how to organize the Internet, depending on which side you take. But here at his company’s headquarters, where he was once CEO, he’s now functionally a product manager. Cohen walks over to his team’s to-do list, which is a collection of paper stuck to a large foam board. “A big chunk of what I do now is re-sorting Post-its,” he says.

Most people know BitTorrent as a technology used to illegally download music and movies, but what few realize is that it’s also a company, a venture-backed startup like any other. “We’re not a wide-eyed startup in that we’ve never done this before; we have our battle wounds,” says Jordy Berson, a manager of BitTorrent’s nine product teams, who’s sitting nearby. Cohen interrupts Berson: “You weren’t even here for all the battles. We went from 60 people down to 10 and back to 60 again.”

That’s only part of the story of BitTorrent’s circular rises and falls. The company has been through five offices and about as many reincarnations. It has launched an iTunes-like content store and an enterprise technology business, licensed its brand to about 30 consumer-electronics manufacturers, and experimented with countless other products–cutting down and building up its staff several times along the way.

Each of BitTorrent’s efforts has revolved around the BitTorrent concept: In 2001, Cohen–a computer programmer with a genius for solving puzzles–created a way to quickly transfer files from one computer to another and called it BitTorrent. It is a method, basically, in which many computers around the world transfer bits of the same file at once, and it has an open-source license that allows anyone to use it however they want (which has often been to pirate content). In 2004, Cohen launched a company by the same name, aiming to make legitimate use of his technology. The company has since evolved to have a larger, loftier mission: to decentralize control of information and “hard-code certain human rights into the Internet, whether that’s the right to privacy or a right to free speech,” says BitTorrent’s current CEO, Eric Klinker.

Revenue now comes to BitTorrent in many forms. Its greatest business success, however, is arguably an income stream that BitTorrent doesn’t like to talk about. BitTorrent (the company) offers free software products (called BitTorrent clients) that people use to download files via BitTorrent (the file-transfer technology). About 800,000 people download its clients a day–two premium,two free–and many don’t notice an option in the free versions asking whether they’d also like to install a toolbar that switches all of their default browser searches and home pages to Yahoo or Ask.com. When a user allows the switch, BitTorrent (the company) gets paid.

This has blessed BitTorrent with a kind of startup immortality–like a teenage zombie from a high-school movie, old and experienced without ever quite growing up. With its clients’ 170 million users and a steady revenue stream, the company can scrap nearly everything and still be guaranteed the revenue needed to reboot. Right now, it’s in growth mode as it tries once again: Its staff has grown from 70 to 150 employees in the past two years, and the company moved into an office space that can accommodate many more. But this spammy, relatively common business model for downloaded products doesn’t live up to Silicon Valley’s romantic idea of disruption, which seems to weigh heavily upon BitTorrent’s executives. “It’s not necessarily a billion-dollar company,” says James Hong, an adviser (and a cofounder of Hot or Not), “and I think they’re always looking around for, What is going to make this a billion-dollar company?”

That’s why, among its many definitions and identities, BitTorrent is ultimately a very curious thing: a gigantic question about what startup success should look like. Most startups don’t achieve their world-changing ambitions, but the lucky ones happily settle for being acquired. BitTorrent seems to be stuck in the middle of all this–profitable, but unable to truly ever capitalize on the revolution it unleashed. “I would argue that BitTorrent did solve a problem, and it did solve the one it set out to solve,” says David Cohen, the CEO of startup accelerator Techstars. What’s supposed to happen next?

BitTorrent, the company, has been working to change the assumption that it enables piracy.

When I visited the company’s San Francisco headquarters last October, it was in the midst of one of its regular hackathon-style brainstorms. Many of BitTorrent’s past initiatives were born at these things: a Dropbox-like product called Sync that now has more than 2 million users and a premium version; a new product for the entertainment industry called Bundles, which has attracted users such as Madonna and Moby; and a BitTorrent chat product that it’s testing on its BitTorrent Labs site. And for the past five years, Cohen has been working on a live-streaming technology that could help outlets such as Ustream and Livestream grow on the cheap: It uses the streaming power of everyone who’s watching, instead of a central server. He invented that one alone.

Cohen doesn’t join the hackathon this time. He stays in his cubicle pod “answering random email,” as he explains. On his desk, next to some mouthwash and a collection of tennis balls, there are a couple of toy puzzles he designed for gift shops in his free time. He lobs one of them, a mess of linked metal pieces, in my direction.

“Take them apart,” he says.

I fiddle with the contraption, but I can’t figure out where it breaks. “I like making puzzles that are very loose, that feel like they’re going to fall apart,” he explains. “But they don’t.”

Past and current coworkers almost uniformly describe BitTorrent’s inventor as “interesting.” Cohen himself tells people he has Asperger’s syndrome. In person, his most prominent attribute, equally admirable and off-putting, is that he doesn’t seem to censor himself. “My teeth hurt, and I got contacted by some random patent trolls this morning,” is the first thing Cohen says when we sit down for a formal interview. “I responded by saying, ‘I fucked your mother and she sucked.’”

Cohen doesn’t like dealing with money and hates having a lot of employees (“It’s empire building,” he says, “and I don’t care about that”). Nor is he much of a salesman. Other startup founders will tell you what they are working on is good for humanity or that it provides a business opportunity that can’t fail. Cohen says he pursued his new live-streaming product “because I had a law of physics calculation that said it could work.”

He doesn’t fit the springy founder archetype, and that may be partly because, back when he invented BitTorrent the technology, he wasn’t necessarily trying to start a company; he was just trying to solve a problem. His friend Andrew Loewenstern liked to record live concerts (a practice bands like Phish encourage), but the files were too big to easily share. Cohen took this as a technical challenge. “Bram loves puzzles,” Loewenstern says. “That’s his thing. He wrote BitTorrent to solve a puzzle.”

Most people’s Internet connections only allow a file to be transferred from one place to another at a plodding pace, in part because computers upload far slower than they download. Transferring a Phish concert is like sending a pickle through a straw. But, Cohen asked, what if many computers all had the same Phish concert, and a new fan wanted to download it? He built a system in which all those computers simultaneously send little bits of the concert to the new fan’s computer, where the bits are then properly assembled. Now there are many straws, each sending little slices of pickle. Files that took days to download suddenly were transferable in hours.

BitTorrent quickly became one of the most popular ways of doing things on the Internet. By 2004, according to one estimate from what at the time was an Internet-traffic management firm called CacheLogic, the protocol accounted for 30% of all Internet traffic. (It has since fallen to about 10%, partly due to new technology aimed at managing BitTorrent bandwidth use and partly because video sites like Netflix and YouTube are using more bandwidth.) Many websites sprang up–with names like the Pirate Bay and ExtraTorrent, which harnessed the BitTorrent technology to enable anyone to transfer just about anything. But there were also myriad other ways to use BitTorrent. Contributors to open-source operating system Linux, for instance, use BitTorrent for distribution. Facebook, Etsy, and others use BitTorrent to update their websites.

“Everybody around Bram,” Loewenstern says, “me, his friends, his family, they were all like, ‘Bram, you need to do something about this and turn it into a business.’” But it wasn’t clear at the time what, exactly, the business potential for BitTorrent was. “The pitch was mostly, ‘We’re BitTorrent,’ ” Cohen says. And what was that? It was hard to say. The BitTorrent technology was–fairly or not–associated with piracy. And its popularity on the web didn’t obviously translate to the corporate world. “It’s like saying, ‘You should buy Google stock because HTTP is 80% of Internet traffic,’ ” says Ashwin Navin, BitTorrent’s cofounder.

But the opportunity was hard to resist. Here was a clearly brilliant inventor with a clearly brilliant idea and a technology with so much engagement that it was quite literally taking over the Internet. In 2005, BitTorrent raised $8.75 million, the first of $40.75 million it would raise over the next three years.

And so began the company’s many identity crises. Back then, its first order of business was not to get sued. A Supreme Court decision that year held peer-to-peer file-sharing companies Grokster and StreamCast Networks liable for their users’ copyright infringement. BitTorrent’s strategy for avoiding a similar fate was to open a store for selling and renting movies. Navin says the idea was a simple matter of protection and reputation: “You won’t be sued by Universal if Universal is working with you.” Warner Bros., MTV, Paramount Pictures, MGM, and dozens of other content providers licensed more than 10,000 videos to the BitTorrent Entertainment Network. But the network didn’t catch on, and in 2008, the company laid off most of the employees who worked on it. It had already begun shifting resources instead to a service called BitTorrent DNA that saved companies money on their bandwidth bills. That was it for Cohen as leader: The company’s board hired a CEO who would focus on building an enterprise sales business, or, as Cohen puts it now, “a guy who wears suits and is viewed as CEO by people who care about such things.”

As the cost of bandwidth plummeted, however, this new concept failed as well. There were more layoffs; the new CEO walked away. DAG Ventures, the lead investor for BitTorrent’s third round of funding, wanted to renegotiate the round. BitTorrent ended up capitulating. It raised its third round, from DAG and others at a discount in the middle of a recession, making early employees’ equity nearly worthless. Navin, who at this point had left the company but was a board member, disagreed and resigned.

That’s when BitTorrent broke its “in case of emergency” glass. Years before, it had worked out its first toolbar deal: Unobservant users would agree to make AskJeeves their default search engine, and AskJeeves would pay BitTorrent for the pleasure. But BitTorrent never pulled the trigger; Cohen and Navin each blame the other for the decision not to move forward. Now BitTorrent needed the money, so it created a similar deal with a company that represents Yahoo and Ask.com (the modern, butler-free incarnation of AskJeeves). “By clicking ‘Accept offer,’ ” reads the small print that many users skip over today, “you are agreeing to: Set Yahoo! as your default search engine, homepage and start-up page in Safari, Firefox and Google Chrome.”

“We basically immediately became profitable,” Cohen says. Eric Klinker, previously the CTO, was appointed to lead the company. He began piecing together yet another line of products. And BitTorrent began considering that its products alone wouldn’t get it to that promised land of billion-dollar success. It had an image problem in need of solving first.

Cohen isn’t particularly interested in making nice in a public-relations kind of way. “I just wanted people to use [the technology],” Cohen says. “And the fact that some people just want to block all new technology just to keep the world as it has been is not my fault or necessarily my problem.” But Cohen isn’t the only voice at BitTorrent anymore. Matt Mason is making this his problem.

Mason is VP of marketing, and is an accomplished salesman of cool. He directed strategy at an ad agency, wrote a book about how innovators can learn from piracy, and was the founding editor of a music magazine in the U.K. He gets up early to go boxing. He is unfailingly polite and talks with his hands. Now he’s become the public face of BitTorrent. When I first meet him in San Francisco, he’s wearing a blue sweater over a blue button-down shirt, with blue jeans and blue shoes. Nearly everyone else in the office is wearing T-shirts, but he looks like a living Gap mannequin. He’s on a tight schedule and is busily eating the tuna off his tuna-melt sandwich. But for me, he recommends the lobster roll: “It’s East Coast good,” he says.

BitTorrent creator Bram Cohen

continues to invent new technologies at BitTorrent while VP of Marketing Matt Mason combats the company’s 10-year-old image problem with a new product for artists.

In 2011, Mason met Klinker at a conference in the south of France. Though he didn’t realize at the time that BitTorrent was a company, he was enthralled by Klinker’s speech. “I was like, this is literally the most important thing happening in the world,” he says. Seven months later, in November 2011, he took a full-time job reshaping BitTorrent’s tattered image.

When Mason started, there was a plan on the table to just rename the company. But he ran an experiment that gave everyone there pause. BitTorrent released a file-sharing product called SoShare and advertised it in two ways–as simply SoShare and as SoShare by BitTorrent. The ads with “BitTorrent” in them attracted five times more clicks. “If you just put the word BitTorrent in front of something people already do on the web,” Mason says, “people understand it means something different.” Changing the name was out.

But Mason saw another opportunity: BitTorrent had begun promoting specific artists’ content in its download clients (it was essentially free advertising, as a way to make nice with artists who were BitTorrent-friendly), and he wanted to expand upon that. So he took a frenzy of trips to L.A. and New York to recruit artists for the project. It wasn’t easy. Few content creators realized BitTorrent the company was a separate entity from BitTorrent the technology. At one dinner, a prominent movie producer opened the conversation with Mason by saying, “I don’t think I like you guys.”

“That’s where the misperception of BitTorrent is,” Mason tells me now. “And until we solve it there with those people, everything else we do doesn’t matter. We could tell Hollywood that we’ve terraformed Mars, and they’d be like, ‘Okay, great, but what are you doing about piracy?’ ”

Mason transformed the experimental promotion of artists into a full-fledged product called BitTorrent Bundles, which makes it easy for artists to put packages of content into the BitTorrent ecosystem. Why would they want to do that? BitTorrent is offering a trade-off: It will work with the artist to create a “bundle” of appealing promotional goodies–maybe a few songs off a new album or a few chapters from a new book, and some bonus material as well. Then BitTorrent will promote the package to its 170 million users. When someone downloads a bundle, they access the free content and then a gate: To get more content, they have to do something–enter an email address or, with an option BitTorrent will begin offering this spring, spend a little cash. The idea is to meet consumers in the middle, giving them something for free in exchange for supporting the artist.

The results are encouraging. Best-selling self-help author Tim Ferriss–desperate after being boycotted by bookstores for using Amazon as a publisher–released the first chapter of his book The 4-Hour Chef in a bundle, with recipes, an audiobook, and live workshop video content available in exchange for a user’s email address. About 210,000 people downloaded the bundle in the first week, and 85,000 of them followed a link to where Amazon sells the full book. Linkin Park used bundles to distribute a free trial of its recording software, StageLight, and soon after saw a 200% increase in the number of upgrades to the paid version. Madonna released a 17-minute film, which was downloaded 2.7 million times.

“Nobody is ready to throw a press conference with BitTorrent and say, ‘These guys are going to save the content industry,’” Mason says. “But there are lots of conversations going on with labels and studios.” Still, talking to the onetime enemy is a delicate affair, and so those conversations are very much off the record. Warner Bros. and Starz declined to comment for this story. Sony directed me to the Motion Picture Association of America. The MPAA declined to comment but sent me a condemning study from anti-piracy services provider NetNames: Among the most popular 10,000 pieces of content downloaded with BitTorrent technology, it claimed, not a single one was being legally offered. (BitTorrent counters that the study wasn’t comprehensive.)

Ask about the definition of success in BitTorrent’s office and you get varied responses. Mason speaks of the company’s progress and rehabilitation, and the ambition to “change the way the whole entertainment industry works.” CEO Klinker speaks of his products: “Success is a probability game. We’re going to stick to our mission, and we’re going to build products most likely to achieve that mission, and we don’t expect all of them to be raging commercial successes.”

And Cohen? “It seems like a silly question,” he says, when I ask if he’s happy with BitTorrent’s direction. “We’re a profitable company.”

Success is hard to define in Silicon Valley, where ambitions and idealism are often set to world-changing levels. Tech execs are trained to think big and to trumpet the absolute goodness of what they’re doing–that TaskRabbit is revolutionizing the workforce, say, or that Uber is a “pro-innovation” champion of consumer choice. “Set a clear, easy-to-understand vision for your company, and make it be a mission people believe in,” wrote Sam Altman, the incoming president of tech accelerator Y Combinator, in a list of startup advice posted to his blog. BitTorrent hasn’t been able to follow that advice: The company was founded after Cohen solved a foundational problem, and so it has sought many other problems and had many successive visions, and thus, has many ways to evaluate its own worth.

But Cohen, as the guy who started it all, may be onto something. What’s wrong with achieving a simple, smaller goal? He built something, then capitalized on it. It’s been a confusing and messy process, and made a lot of enemies, but the business he started also steadily employs roughly 150 people. His technology improved the Internet in a meaningful way. Maybe nothing loftier will be achieved. Maybe no acquisition will follow. “I do think that in some sense, being alive and continuing to experiment and support yourself is a success,” says Techstars’ David Cohen. “You’ve built a sustainable business. Just because it’s not WhatsApp [doesn’t mean you’ve failed].”
So the experimentation continues.

Along the highway that connects Silicon Valley to San Francisco, a billboard appeared this past September that seemed to taunt rather than entice tech enthusiasts: “Your data should belong to the NSA,” it said. Similar signs popped up the same week in Los Angeles and New York: “The Internet should be regulated,” some said; “Artists need to play by the rules,” said others.

Passersby took to social media and blogs with guesses about who could be behind the anonymous ads. Was it the street artist Banksy, who was in the middle of a monthlong residency in New York at the time? Was it Michael Arrington, the founder of TechCrunch, who simply seems to enjoy antagonistic gestures? A month later, the answer appeared on the billboards themselves. Corrected to read, “Your data should belong to you, the Internet should be people-powered, and artists need options,” they now featured the BitTorrent logo.

This is BitTorrent’s newest identity. Products built using decentralized technology like BitTorrent, such as the company’s chat and Dropbox-like Sync, don’t store data on central servers, so user content is theoretically less susceptible to mass surveillance. (It’s still easy to see who downloaded what using BitTorrent’s protocol; the data just isn’t stored in one easy-to-collect place in the cloud.) It was Mason’s handiwork. “There aren’t many other tech companies that could put up an anti–NSA billboard right now or say the Internet should be people-powered,” he says. “It was a way for us to set up what we stand against and show people we’re on their side.”

Maybe it will catch on. If not, BitTorrent will try again.

About the author

Sarah Kessler is a senior writer at Fast Company, where she writes about the on-demand/gig/sharing "economies" and the future of work.



More Stories