How the Internet Happened

by Brian McCullough

  This was the closest thing to an existential crisis Facebook had ever faced. The history of social networking had shown that users were fickle; they would flock to whatever service best suited their needs at the moment. If you pissed off your users, they would leave you. The reasons Friendster had been abandoned were largely technical, but sites could be brought low by basic design changes as well. A few years after the News Feed brouhaha, Digg would redesign its site and change its voting algorithms in a way that so angered users that they fled, en masse, to a Digg competitor named Reddit. To this day, Reddit is known as the “front page of the Internet,” the birthplace of memes and viral culture, while Digg, though still around, is nowhere near as relevant or well trafficked.

  So, with the News Feed backlash, panic set in at Facebook HQ. High-level meetings were held among the Facebook brain trust over whether or not to backtrack and shut off the News Feed. Zuckerberg himself quickly penned a note to users, “Calm down. Breathe. We hear you.” Privacy controls were hastily coded up to give users better control over what showed up on the Feed and what didn’t. But the News Feed was never shut down, even temporarily, because, again, Zuckerberg was watching user behavior and, despite the ruckus, he could see that people were actually using the News Feed as he had intended. In August, before the News Feed, Facebook users viewed 12 billion pages. In October, après News Feed, pageviews were 22 billion.66 People might claim to hate the feature, but Zuckerberg could see they couldn’t stop using it. In fact, the proliferation of anti–News Feed protests was tangible proof that the new feature was working as designed. The whole point of the News Feed had been to surface things happening you might want to know about, he told Fortune reporter David Kirkpatrick at the time. “One thing it surfaced was the existence of these anti-feed groups.”67 The News Feed itself had enabled its own backlash to spring up.

  The anger blew over eventually, and the News Feed went on to become the core feature of Facebook. But it still caused a very real crisis in confidence at a crucial and uncertain time. “If [News Feed] didn’t work,” Chris Cox says, “it confounded [Zuckerberg’s] whole theory about why people were interested in Facebook. If News Feed wasn’t right, he felt we shouldn’t even be doing [Facebook itself].”68

  It didn’t help that the News Feed near-fiasco came on the heels of a less publicized but no less demoralizing failure from earlier in the summer. When Facebook’s work networks were launched, they barely got any attention. Only on army bases, and among U.S. military users, had the workplace networks taken off. But then, military folk were generally the same college-age cohort that Facebook had always been successful with. Adults didn’t seem to be interested in the service at all.

  The News Feed experience shook Zuckerberg’s core faith in what Facebook was all about. And after the failure of work networks, a bigger question hung heavy in the air: was Facebook really just for kids after all? If so, then Zuckerberg’s great insight, that his social graph was a useful thing for everyone on the planet, was mistaken. “It was the most wrong he’d ever been at Facebook, and the first time he’d ever been wrong in a big way,” early Facebook executive Matt Cohler said of this period of doubt.69 If Zuckerberg was wrong about these things, had he also fundamentally misjudged the big, world-changing value of Facebook to begin with? In that case, maybe a $1 billion sale wasn’t such a bad outcome after all. They had already conquered the high school market. Myspace still had a lead in the overall twenty-something demographic. If older users couldn’t be enticed to join, there wasn’t any more low-hanging fruit to be had in terms of harvesting growth.

  And it was at this exact moment, in September 2006, that Yahoo came back and renewed its $1 billion all-cash offer. Yahoo’s lawyers did due diligence on Facebook’s finances and operations, and an acquisition was agreed to in principle. Given the stumbles of the past few months, nearly everyone was now in favor of a sale—especially the VC investors, but plenty of rank-and-file Facebook employees as well.

  Everyone, that is, except for Mark Zuckerberg. And even he was beginning to waffle.

  “We almost took the offer,” Sean Parker would later say.70 It was seemingly the only time the pressure to sell got to be too much for even Zuckerberg to resist.

  But before agreeing to sell, Zuckerberg wanted to take one last crack at opening Facebook up to everyone. Once more, he played for time, dragging his feet on the acquisition talks, taking meeting after meeting but not actually pulling the trigger on the Yahoo deal. He wanted to see if his gut instincts about Facebook were right.

  Perhaps—perhaps the work groups had failed because they were the wrong paradigm. Maybe Facebook had used its tried-and-true network-by-network expansion trick one time too many. Maybe explicit networks were less important outside of a school setting. The people who graduated college but still continued to use Facebook just took the network with them, even when they moved away from campus. Perhaps the thing to do was just throw registration wide open and let everyone in. That way users could grow their networks organically.

  The engineers borrowed an idea from Sean Parker. Plaxo had grown by searching users’ existing address books and email programs to invite people to join and make connections. An “Address Book Importer” was designed to go into your Hotmail or Gmail account and search for other users who were already on Facebook. That way, new users would be greeted with a slew of people they already knew when they signed up and needed to begin populating their network. The importer would serve up friend connections on a platter, and anyone not on the service could be invited to join via the same mechanism.

  It was one last roll of the dice. One last gamble, where failure still meant $1 billion and success meant—well, who knew?

  ■

  OPEN REGISTRATION WAS LAUNCHED on September 26, 2006, mere weeks after the News Feed debacle. Prior to open registration, new users were joining at a rate of about 20,000 a day. A few weeks after opening up Facebook to everyone, that number had changed to 50,000 a day, and rising.71 Growth in Facebook’s user numbers began to look like a hockey stick going only steeply upward. Over the next year, Facebook would rocket past 25 million registered users, and around 6 million of those would be older-than-college-age users; 200,000 of those would even be people over age sixty-five.72 If you were a postcollege adult during this period, you might remember this moment. One day, Facebook was just a thing you had heard of. The next day, everyone you knew was on it. Some day after that, your mother and even your grandmother were members.

  The one personal anecdote I’ll share in this book: that summer of 2006 was my ten-year high school reunion. It was an important event. Many of my classmates had lost touch with each other. There were a lot of “Wow! What happened to you?” conversations. And then, just a few months after we got together, open registration happened, and we all found each other again on Facebook. Soon we were all even connected with classmates who hadn’t been able to make the reunion.

  Ten years later, our twenty-year high school reunion in 2016 was less of an event. It was more of “Hey, I saw the photo of your new car yesterday” than it was “Where have you been?” After all, thanks to Facebook, I now get updates about everyone on an hourly basis. I know that my senior-year chemistry lab partner just got back from a trip to China and that the oldest child of the girl I kissed in sophomore year just broke his arm skateboarding. There is a very clear demarcation point to my social life between pre-Facebook times and post-Facebook times, and it felt like the change happened overnight.

  ■

  IT SORT OF DID HAPPEN OVERNIGHT. From its launch in 2004 until open registration in 2006, Facebook grew to around 8 million users.73 One year after open registration, Facebook had 50 million active users.74 By the end of 2008, there were 145 million people on the service, 70% of them outside the United States.75 The next year, there were 350 million users in 180 countries. After open registration, the social-networking wars were over. Myspace, and every other social network, would become distant memories.

 
; It turned out that Mark Zuckerberg was right. Connecting everyone together—almost the original premise of the web itself—was an incredibly useful and valuable thing indeed. Zuckerberg is the twenty-three-year-old who turned down a billion dollars because he thought he was sitting on an idea that was even bigger. The gamble has paid off (at the time of this writing) to the tune of a nearly half a trillion dollars in market value. It helped that advertising against everyone’s personal lives also proved to be lucrative, and that the reverse-chronological scrolling mechanism of the News Feed proved to be perfectly suited for the coming age of mobile computing. But none of that would have been possible had Zuckerberg not matured into the sort of businessman who could make such a gamble. The fact that he did is the entrepreneurial story of our age.

  16

  THE RISE OF MOBILE

  Palm, BlackBerry and Smartphones

  In the technology world, the ultimate success of a new idea is very much dependent on timing. Even great ideas that are quite obviously “the next big thing” can fail to deliver on their promise because the underlying technology or infrastructure isn’t mature enough yet. Streaming video was supposed to be big, going back to the days of Real|Audio and Broadcast.com, but it took the example of Napster and the advent of broadband Internet connections before YouTube could take off. SixDegrees couldn’t succeed because it was birthed in a world before ubiquitous digital cameras. Facebook got the timing right on that detail, but it also cracked the social-networking code by achieving critical mass just as another key technology was having its breakthrough moment.

  For many long years, mobile computing was an idea before its time. Dozens of attempts to jump-start mobile computing as an industry crashed and burned without gaining widespread adoption. After the PC revolution and just prior to the dot-com era, there was a brief fad in Silicon Valley for handheld computers. It was the logical next step: once there was a computer on every desk, why not put one in every pocket? In the late 1980s and early 1990s, there was a minibubble as investors rushed to fund dozens of handheld, mostly pen-based, computer startups, both on the software and the hardware side. GO Corp. burned through $75 million dollars in VC money in an attempt to become the Microsoft of handhelds by creating the operating system standard for pen computing. A company called GeoWorks attempted something similar with its GEOS. General Magic was an Apple spin-out that Pierre Omidyar of eBay; Tony Fadell, the father of the iPod; and Andy Rubin, the inventor of the Android operating system, all worked at before going on to fame and fortune elsewhere.

  Before it was spun out, General Magic was one of two top-secret research-and-development teams1 inside Apple Computer. The other team, which remained in-house, was called Newton, and it would be responsible for the highest-profile early handheld computing device. As the 1980s turned into the ’90s, the Newton team was working on a tablet computer the size of an eight-and-half-by-eleven-inch sheet of paper. This experimental device weighed about eight pounds but was only three-quarters of an inch thick. Named Figaro, the machine was navigated using a stylus on a grayscale screen, had three processors, an internal hard drive and wireless networking and got about ten hours of battery life. Oh, and it cost about $8,000 to produce.2 Per device.

  In early 1991, a young Apple marketing executive named Michael Tchao convinced Apple’s then-CEO, John Sculley, to switch gears and have the Newton team work on a smaller, ultraportable computer—one that could fit comfortably in a person’s palm. Sculley became an evangelist for the idea of a near-pocketable computer, a category of devices he termed personal digital assistants, or PDAs. At the 1992 Consumer Electronics Show, Sculley declared that there would soon be a “$3.5 trillion” market for such devices.3

  The result of this strategic pivot, the Newton MessagePad, was released to the public on August 2, 1993. It cost $699, ran on four AAA batteries and weighed 0.9 pounds. But at 7.24 inches by 4.50 inches (about the size of a VHS cassette) it was hardly pocketable, except in the most generously sized pockets.4 With optional add-ons, you could send faxes and (eventually) email using a wired modem. But the main features of the Newton were its productivity apps, including a calendar, address book, to-do list and notepad. It had no keyboard, instead boasting a touchscreen that you interacted with using the included stylus. The intention was, you would write on the Newton just as you would if you were writing on a piece of paper. The software would interpret your handwriting and turn it into on-screen text.

  Or, at least, it was supposed to. The Newton was ultimately done in by its notoriously flaky software, which, more often than people could tolerate, simply refused to recognize what had been written. Oddly enough, the longer the word, the better the software was at translating, because longer words gave the handwriting interpreter more information to work with. The Newton struggled primarily with shorter, monosyllabic works like “or” and “the.”5

  The Newton’s software was supposed to learn your handwriting over time, but PC Week complained that, “The Newton is almost worthless . . . basically shelfware. After three weeks, it still couldn’t consistently differentiate my 1’s from my t’s.”6 Other reviews were just as scathing. “Apple promised too much and failed to deliver a useful device,” wrote the New York Times.7 In a classic example of a rolling PR catastrophe, after the Newton came out, the comic strip Doonesbury spent a week turning the Newton’s handwriting recognition foibles into a national joke.

  Apple had expected to sell 1 million Newtons in the first year. It sold only 85,000.8 Subsequent models would improve immeasurably, especially the second-generation device that was Jony Ive’s first assignment after being hired at Apple. But it was too late. In the court of public opinion, the Newton could never overcome its poor reputation.

  ■

  THE NEWTON’S HIGH-PROFILE failure took the entire nascent handheld computing market down with it. Most of the handheld startups went out of business in the coming years, just as Silicon Valley was turning its attention to the web. One of the handhelds that made it to market, only to be dragged down in the Newton’s wake, was the Zoomer, a $700, one-pound pen computer that debuted in October of 1993, selling only 60,000 units before being discontinued.9 But even a firsthand brush with failure could not kill the dream of a “computer you could carry in your pocket,” not for Jeff Hawkins, the inventor of the Zoomer. Hawkins had founded Palm Computing in January 1992 to produce the Zoomer, and even after its first product failed in the marketplace, Hawkins and a small band of Palm loyalists merely went back to the drawing board and began sketching out a follow-up device.

  Hawkins had a hunch that handhelds had attempted to do too much, had been too complex, too ambitious. He intuited that people didn’t necessarily want a second computer, they wanted an accessory to their existing computer. So, he focused on only a few key use cases for his new device: a calendar, an address book and a memo pad. These applications would be designed to sync to regular computers when the device was connected by wire; when out in the “wild,” as it were, the device would stick to its primary, simple task: helping the user stay organized.

  Hawkins began carrying a rectangular piece of balsa wood, about the size of a deck of cards, around Palm Computing’s offices. With this dummy mock-up, Hawkins tested out the ideal dimensions that would allow a handheld device to be useful in everyday situations. The resulting product would be known as a PalmPilot (though it had various names due to branding and trademark issues over the years). By sticking to Hawkins’s ethos of simplicity, not only was the Pilot eminently pocketable (it was about a third the size of the Newton and weighed 5.5 ounces); it could also hew to a $300 price point, thereby making it seem like a logical desktop or laptop accessory.

  Palm would sell 1 million Pilot units in eighteen months on the market, thereby becoming the fastest-selling computing device in history.10 It was still a pen-based gadget—there was no physical keyboard—but Hawkins had solved the handwriting input issue that had beguiled the Newton by creating a single-stroke shorthand alphabet known as �
�graffiti.” This improvised input language worked well and the Pilot proved useful, especially to businesspeople on the go, with a simple plug-in-and-sync interface, much akin to what would later become commonplace with the iPod and iTunes system. By 2001, Palm had sold 21 million of these pocket computers and secured a 70% market share of a reborn PDA market.11

  In Canada, another small company took notice of the rebirth of the pocket computer and decided to come at the market from a different angle. If Jeff Hawkins focused on the simplicity of productivity and organization while on the go, Mike Lazaridis, the founder of Research In Motion (RIM), focused on communicating while on the go. In 1996, RIM launched the Inter@ctive Pager, a two-way wireless messaging device. Initially, it was just a glorified pager. But Lazaridis and the RIM engineers concocted clever ways to hook into personal and corporate email systems and eventually, RIM was delivering, essentially, email in your pocket. The first Inter@ctive Pager, the 900, and its successor, the 950, released in September of 1998, shared the Palm ethos of simplicity, pocketability and utility on the go. Measuring 2.5 ✕ 3.5 inches and weighing 4.5 ounces, the RIM pagers mimicked the PalmPilots in their form.12 “Everyone else was trying to add a radio to a PDA,” recalled Dr. Peter Edmonson, RIM’s chief radio engineer. “Whereas Mike’s mindset was how to add a PDA to a radio.”13

  RIM’s devices were designed to be “always online” as opposed to syncing to a computer occasionally, as the PalmPilots were designed to do. Email was “pushed” to RIM’s gadgets over the wireless network, so you didn’t have to plug in to find your messages; your messages found you, wherever you happened to be. When you got something new in your inbox, the device would buzz and a red LED would indicate that you had a new message to read. Because RIM had previous experience working with radios and wireless networks, its pagers were fast and incredibly energy-efficient. The 950 could last for three weeks on a single AA battery. And RIM didn’t mess with the touchscreen technology that Palm was so married to. Instead, RIM innovated tiny, fully functional keyboards designed to be used with one’s thumbs. “For me, it was all about keyboards,” Lazaridis has said. “Jeff [Hawkins] went off and did touch screens. I went off and tried to develop something with a keyboard.”14