|THE INSIDE STORY OF BITTORRENT’S BIZARRE COLLAPSE|
Jessi Hempel - Backchannel
Last April , a pair of cousins named Bob Delamar and Jeremy Johnson became co-CEOs of BitTorrent. Delamar was a bearded Canadian Japanophile in his early forties; Johnson a network engineer from San Diego. Through an unusual financial arrangement, they represented a four-person group that had recently come to own a controlling stake in the company, and they had a plan to turn BitTorrent into, as Delamar was fond of saying publicly, “the next Netflix.” BitTorrent had already tried to be the next Netflix, starting long before Netflix had become the next Netflix. The company was founded in 2004 by Bram Cohen, inventor of the open-source protocol that lent the startup its name, and Ashwin Navin. BitTorrent — the protocol — was a genius way to transmit large amounts of information over the net by breaking it into small chunks, sending it through a peer-to-peer network, and reassembling it. BitTorrent — the company — got started on the assumption that Cohen was brilliant. He’d invented one of the web’s most fundamental tools, and surely there was a business to be made from it.
But from the start, BitTorrent had a branding problem — pirates used it to share movies illegally, making it the Napster of entertainment. Because the protocol was open-source, BitTorrent (the company) couldn’t stop the pirates. For 12 years, BitTorrent’s investors, executives and founders attempted to figure out many money-making strategies, including both enterprise software and entertainment businesses, while convincing us all that, sure, people might use the BitTorrent protocol to conduct illegal activity, but BitTorrent was just a tool — a really great tool you can use for really great things!
They’re right: 170 million people used the protocol every month, according to the company’s website. Facebook and Twitter use it to distribute updates to their servers. Florida State University has used it to distribute large scientific datasets to its researchers. Blizzard Entertainment has used BitTorrent to let players download World of Warcraft. The company’s site boasts that the protocol moves as much as 40 percent of the world’s Internet traffic each day.
Jessi Hempel is Backchannel's editorial director.
Sign up to get Backchannel's weekly newsletter.
But transforming this technology into any kind of business has proved elusive. By last spring, BitTorrent had already endeavored to become a media company, twice. There was BitTorrent Entertainment Network, launched in 2007, which was a storefront for movies and music that made no money and shut down a year later. And then there was the BitTorrent Bundle, launched in 2013, which was a competitor to iTunes and Amazon that let artists distribute their work directly to fans at a fraction the cost. In 2014, the company even announced plans to produce its own original series, a scifi show called Children of the Machine. But by early the next year, BitTorrent had given up on this strategy, too.
Some startups are born lucky. By the chance of their timing, their technology, or the individuals who helm them, they experience Facebook-size success. Others fail quickly. There is luck in this, too — in an immediate, concise conclusion. Far more startups, having raised funding on the merits of an idea and a team, plod along for years or even decades, constantly casting about for the idea or customer or partnership that will transform them. Their investors are patient, and then exhausted, and then checked out, and then impatient. Their executives change, and then change again. The founders leave, or they hang on in hopes the company they conceived will somehow eventually prove itself. They are zombie startups.
Such is the case with BitTorrent. It has remained a technology in search of a business for a dozen years. Then last year, Delamar and Johnson arrived with plans to save it once and for all. Instead, they squandered millions on failed schemes, putting the company on course for collapse.
I stumbled across this story while reporting Backchannel’s weekly Follow-up Friday piece, in which we step out of the knee-jerk news cycle to follow up on announcements and news events from previous years. I reached out to discover what had happened to Children of the Machine, the original series for which BitTorrent received accolades for announcing two years ago. When the company didn’t respond, I began asking others.
BitTorrent doesn’t want to talk about what happened last year. It made no executive available to answer questions. I pieced together the following narrative by speaking with current and former employees, investors and artists. Consider it a morality tale for discordant investors and entrepreneurs. It’s the story of the most recent dramatic and strange chapter in the life of one venture-backed company that has failed to succeed, but also hasn’t failed.
As a child on Manhattan’s Upper West Side, Bram Cohen was smart, introverted, and strange. “I knew I was weird,” Cohen once told FORTUNE, explaining that he got frustrated trying to interact with other people. “I can really remember lots of stories in my life — things that it’s really obvious to me now what was going on, but I didn’t realize it back then because I didn’t understand people very well.” He graduated from Stuyvesant High School. But for all of his ability to focus, his grades were dismal. He attended the University of Buffalo, dropping out after two years.
Cohen has Asperger’s Syndrome, a condition about which he has always been very public. He disclosed his condition to an early investor, for example, during one of their earliest fundraising meetings. “It’s one of the first things he tells most people,” the investor told Bloomberg BusinessWeek in a 2008 profile. As a result, he’s not a handshaker. He doesn’t like wearing shoes. He’s not one for making small-talk.
In his mid 20s, having worked a string of dot-com jobs, Cohen spent the better part of nine months hunched over a Dell keyboard at his dining room table, consumed by a puzzle he could only solve by writing code and more code. He lived off his savings, and later credit cards. He felt certain he could figure out how to solve a puzzle that had stumped programmers since the start of the web — how to transfer massive files. The result, of course, was the open-source protocol BitTorrent.
In 2004, Cohen partnered with his younger brother, Ross Cohen, and Ashwin Navin, an alum of Goldman Sachs and Yahoo, to attempt to create a business around the protocol. They raised $8.75 million from Doll Capital Management (DCM). An early business plan was to establish a marketplace, like eBay, for creators to sell bandwidth-intensive content to consumers. They’d make money off it either through advertising or by charging these sellers a fee. The venture firm Accel led the company’s next round, in December 2006.
From the start, the company had personnel issues. Early on, Cohen’s brother, who had been in charge of the engineers, left. In 2007, Cohen ceded the CEO role to a short-lived outsider, moving into the newly created role of Chief Scientist (a title he has kept). In 2008, Eric Klinker, who was then chief technology officer, became BitTorrent’s CEO. Klinker possessed a rare combination of traits — he had the people skills to run the company, and he was sharp enough technically to win Cohen’s respect. (This was a particularly high bar.)
The original business idea didn’t take off, and for years the company cast about for promising alternatives. In 2008, having taken a third round of financing, the company admitted the business wasn’t “gaining significant traction” and agreed to recapitalize. It returned the $17 million to investors and instead raised just $7 million — from the same investors — at a significantly reduced valuation. It was a sign the company was in trouble. Navin left. And still, the company tried to make a go of it.
So went the life of BitTorrent. The company was headquartered in a gray office complex in San Francisco’s SOMA district. The executives tried strategies, hired people, experienced failures, and laid people off at regular intervals. A TechCrunch post from 2010 begins, “Hmm, BitTorrent…that’s still around?”
The latest chapter of BitTorrent’s saga begins in earnest in 2015. By then, many of the company’s executives and directors were exhausted. They still couldn’t agree on a path forward for the company. Some people believed it should double down on its technical business, building products people loved. They’d developed a product called Sync, for example, which was a decentralized version of Dropbox. Others wanted it to be an entertainment company, striking deals to send content to those people. With no focus, the company had reached an impasse. Earlier that year, BitTorrent had laid off nearly a third of its 150 employees. That’s when Accel’s Ping Li decided he wanted out. He’d been invested in BitTorrent since 2006, when he led a $20 million round of financing. Back then, he’d been excited about the company’s potential. But after a decade in which it had failed to hatch a venture-size business, he couldn’t see a path forward. Says Li, “We couldn’t get excited by any of the plans after ten years. We thought the best way to support them is to let them do what they do.” Also, BitTorrent was among the last outstanding investments in the Accel fund that had had an early stake in Facebook and Dropbox, among others — possibly the best performing venture fund of all times — and the firm was looking to wrap it up.
That’s when a group of investors offered to step in. They were familiar with BitTorrent because one of them, Jeremy Johnson, had been friendly with Klinker; the pair had worked together starting back in the late 1990s at the internet service provider Excite@Home, and had gone on to work on an Accel-backed routing startup together. By fall, the investors had obtained Accel’s stake in BitTorrent.
By venture norms, this was an unusual transaction. Here’s how it worked: Johnson and his cousin, Robert Delamar, teamed with two others to start an investment company called DJS Acquisitions. They had no money to offer up front, but they volunteered a $10 million promissory note in exchange for Accel’s stake in BitTorrent as well as DAG’s remaining stake in the company. (DAG was a minority shareholder, having first invested also in 2008.) The plan was that DJS would repay the note in a year.
It’s uncommon for an investment firm to exchange its shares for a promissory note. Why did this make sense for Accel and for BitTorrent? Well, for one, the DJS team articulated a plan for transforming BitTorrent into an entertainment company. Sure, it hadn’t worked before, but they showed up with new blood and new enthusiasm. Beyond that, it wasn’t clear Accel had other options. While some insiders said that Cohen had tried to buy parts of the company back himself, Accel’s Li didn’t feel there were other reasonable options on the table.
Regardless, the resulting transaction gave the DJS team, which had not actually invested any capital yet, a good deal of power in the company. DJS inherited two of the company’s five occupied board seats, replacing Ping and the partner from DAG with Johnson and Delamar. It owned more than 50 percent of the company’s preferred shares, according to four people with direct knowledge of the company’s corporate structure. In other words, DJS was in control.
The four members of the DJS team had eclectic backgrounds. Two had come up in engineering: Johnson and Raj Vaswani, cofounder of Silver Spring Networks. The other two are in business together at a Vancouver-based startup called Pacific Future Energy. Its goal is to build an oil refinery in British Columbia. Delamar, a lawyer by training, was chief executive of this endeavor and is now a senior advisor, and Samer Salameh is executive chairman. Within a few months of their arrival, Klinker resigned as CEO. The board appointed Delamar and Johnson as co-CEOs, and they were free to pursue their strategy of turning BitTorrent into a Hollywood behemoth. By June, BitTorrent had divorced its media and enterprise businesses, spinning its Sync product into a standalone company called Resilio. Klinker runs it. Today, Resilio offers freemium software for companies.
Meanwhile, Johnson and Delamar moved quickly to realize what they believed to be BitTorrent’s media opportunity. Delamar made plans to open an office in Los Angeles, and began commuting between LA and Vancouver, where he lived in a two-bedroom rental in the Shangri-La Hotel building. Meanwhile, Johnson opened an engineering office near his San Diego home. (Neither of them made it regularly to the company’s San Francisco headquarters, in a gray office complex just South of Market Street.)
They went on a hiring tear, boosting headcount by 26 percent between January and June, with most of the new hires in marketing and sales. They also brought in some of their own people as senior executives, a few of whom remained employed at Pacific Future Energy at the same time. Salameh, who is currently CEO and executive chairman of PFE, was paid a consulting fee by BitTorrent that totaled $154,000. Delamar, who remains a senior advisor to PFE, also hired Jeremy Friesen, who is PFE’s chief investment officer, as executive vice president of corporate development; Friesen worked for both companies simultaneously.
The pair moved quickly — at great expense — to spread the word in Hollywood and beyond that BitTorrent was a smart option for distributing movies and music, one that allowed artists to be in control of their distribution and had the potential to reach large audiences. They hired Missy Laney, who had managed Sundance Institute’s Artist Services Program, to help woo filmmakers. They relaunched their platform intended to let artists distribute their work directly to fans, calling it BitTorrent Now. They hired the son of a former CNN anchor to start an online news outlet. They launched the Discovery Fund, promising up to $100,000 in grants to 25 aspiring artists. They even paid a female motocross big truck driver, reportedly a friend of Johnson’s, $50,000 to plaster the company logo across the side of her truck.
Even as BitTorrent’s ad revenue was apparently declining, Delamar spent much of his time trying to convince Hollywood producers that BitTorrent could deliver massive audiences and profits for their creative work. In an August email to X-Men producer Tom DeSanto that he shared with the entire company, Delamar suggested a plan to generate a billion dollars for DeSanto’s next project by releasing it via BitTorrent, writing, “Our goal is to do something that has never been done before here with you.” In an email, DeSanto told me the talks didn’t go anywhere, writing: “Bob was very excited by my ideas but I have no plans right now to partner with bit torrent.”
By the end of the summer, it had become clear the strategy wasn’t working. The pair blew through more than a third of the company’s existing cash reserve, while revenues declined. BitTorrent had, for several years, maintained cash reserves of $33 million, give or take a few hundred thousand, according to financial documents shared with the board. By last July, the company had $14.9 million in cash, and forecasted ending the year with just more than $8 million in cash. The company had spent $10.1 million in the first six months of the year.
Amid all of these efforts Cohen had little sway — and little interaction with the rest of people at the company he had created to make something of his invention. His equity had been so diluted that he had little voice; the professional investors controlled 70 percent of BitTorrent. And within the company itself, Cohen had no direct reports. For the last few years, he has poured his energy into BitTorrent Live, a technically complex piece of software that allows people to broadcast live directly to viewers. Quietly, over the summer, after several years of development, the company released the app in beta.
In October 2016, a year after DJS struck its deal with Accel, the promissory note came due. DJS reportedly was unable to pay. DCM’s David Chao, the remaining venture investor, reportedly stepped in to pay the note, assuming control of their shares — and affording three board seats to DCM. BitTorrent fired its newly impotent co-CEOs. Today, the company’s chief financial officer, Dipak Joshi, is interim CEO. Both Delamar and Johnson have left the company. BitTorrent has shuttered its LA production studio and San Diego office, and laid off a larger number of its staffers. The Discovery Fund that announced grants to artists in August has finally sent an email to all applicants saying the program has been suspended. (“Sorry, Discovery fund has been scrapped out.”)
It’s unclear what’s ahead for the company. I did, however, finally track down the creator of Children of the Machine, Marco Weber, who told me he has finished writing the series and is currently shopping it in a more traditional manner. Anxious fans may one day get to see it after all, though likely not on BitTorrent.
Nearly everyone to whom I spoke had a different perspective on what had gone wrong at the startup. Infighting. Profligate spending. Strategic mistakes. But to a person, every last one agreed on one thing: the technology that Cohen invented was brilliant. Said one person, “It’s a testament to Bram’s genius that no one has yet built a better trap for moving this big data over bad networks.”
Perhaps the lesson here is that sometimes technologies are not products. And they’re not companies. They’re just damn good technologies. Vint Cerf did not land a Google-size fortune for having helped invent the TCP/IP protocols that power the Internet (though he did get the U.S. National Medal of Technology). What’s more, to be successful, a startup requires both a great idea for a product or service, and a great idea for how to make money off of it. One without the other will fail.
Then again, like so many other zombie startups littering Silicon Valley, BitTorrent is not dead yet. Just before the holidays, Cohen’s BitTorrent Live app debuted in the app store.
--------------------------------- END ---------------------------------