فصل 04

4

ELON’S FIRST START-UP

IN THE SUMMER OF 1994, Musk and his brother, Kimbal, took their first steps toward becoming honest-to-God Americans. They set off on a road trip across the country.

Kimbal had been working as a franchisee for College Pro Painters and done well for himself, running what amounted to a small business. He sold off his part of the franchise and pooled the money with what Musk had on hand to buy a beat-up 1970s BMW 320i. The brothers began their trip near San Francisco in August, as temperatures in California soared. The first part of the drive took them down to Needles, a city in the Mojave Desert. There they experienced the sweaty thrill of 120-degree weather in a car with no air-conditioning and learned to love pit stops at Carl’s Jr. burger joints, where they spent hours recuperating in the cold.

The trip provided plenty of time for your typical twenty-something hijinks and raging capitalist daydreaming. The Web had just started to become accessible to the public thanks to the rise of directory sites like Yahoo! and tools like Netscape’s browser. The brothers were tuned in to the Internet and thought they might like to start a company together doing something on the Web. From California to Colorado, Wyoming, South Dakota, and Illinois, they took turns driving, brainstorming, and talking shit before heading back east to get Musk to school that fall. The best idea to arise from the journey was an online network for doctors. This wasn’t meant to be something as ambitious as electronic health records but more of a system for physicians to exchange information and collaborate. “It seemed like the medical industry was one that could be disrupted,” Kimbal said. “I went to work on a business plan and the sales and marketing side of it later, but it didn’t fly. We didn’t love it.”

Musk had spent the earlier part of that summer in Silicon Valley, holding down a pair of internships. By day, he worked at Pinnacle Research Institute. Based in Los Gatos, Pinnacle was a much-ballyhooed start-up with a team of scientists exploring ways in which ultracapacitors could be used as a revolutionary fuel source in electric and hybrid vehicles. The work also veered—at least conceptually—into more bizarre territory. Musk could talk at length about how ultracapacitors might be used to build laser-based sidearms in the tradition of Star Wars and just about any other futuristic film. The laser guns would release rounds of enormous energy, and then the shooter would replace an ultracapacitor at the base of the gun, much like swapping out a clip of bullets, and start blasting away again. Ultracapacitors also looked promising as the power supplies for missiles. They were more resilient than batteries under the mechanical stresses of a launch and would hold a more consistent charge over long periods of time. Musk fell in love with the work at Pinnacle and began using it as the basis for some of his business plan experiments at Penn and for his industrialist fantasies.

In the evenings, Musk headed to Rocket Science Games, a start-up based in Palo Alto that wanted to create the most advanced video games ever made by moving them off cartridges and onto CDs that could hold more information. The CDs would in theory allow them to bring Hollywood-style storytelling and production quality to the games. A team of budding all-stars who were a mix of engineers and film people was assembled to pull off the work. Tony Fadell, who would later drive much of the development of both the iPod and iPhone at Apple, worked at Rocket Science, as did the guys who developed the QuickTime multimedia software for Apple. They also had people who worked on the original Star Wars effects at Industrial Light & Magic and some who did games at LucasArts Entertainment. Rocket Science gave Musk a flavor for what Silicon Valley had to offer both from a talent and culture perspective. There were people working at the office twenty-four hours a day, and they didn’t think it at all odd that Musk would turn up around 5 P.M. every evening to start his second job. “We brought him in to write some very menial low-level code,” said Peter Barrett, an Australian engineer who helped start the company. “He was completely unflappable. After a short while, I don’t think anyone was giving him any direction, and he ended up making what he wanted to make.”

Specifically, Musk had been asked to write the drivers that would let joysticks and mice communicate with various computers and games. Drivers are the same types of annoying files that you have to install to get a printer or camera working with a home computer—true grunt work. A self-taught programmer, Musk fancied himself quite good at coding and assigned himself to more ambitious jobs. “I was basically trying to figure out how you could multitask stuff, so you could read video from a CD, while running a game at the same time,” Musk said. “At the time, you could do one or the other. It was this complicated bit of assembly programming.” Complicated indeed. Musk had to issue commands that spoke directly to a computer’s main microprocessor and fiddled with the most basic functions that made the machine work. Bruce Leak, the former lead engineer behind Apple’s QuickTime, had overseen the hiring of Musk and marveled at his ability to pull all-nighters. “He had boundless energy,” Leak said. “Kids these days have no idea about hardware or how stuff works, but he had a PC hacker background and was not afraid to just go figure things out.”

Musk found in Silicon Valley a wealth of the opportunity he’d been seeking and a place equal to his ambitions. He would return two summers in a row and then bolt west permanently after graduating with dual degrees from Penn. He initially intended to pursue a doctorate in materials science and physics at Stanford and to advance the work he’d done at Pinnacle on ultracapacitors. As the story goes, Musk dropped out of Stanford after two days, finding the Internet’s call irresistible. He talked Kimbal into moving to Silicon Valley as well, so they could conquer the Web together.

The first inklings of a viable Internet business had come to Musk during his internships. A salesperson from the Yellow Pages had come into one of the start-up offices. He tried to sell the idea of an online listing to complement the regular listing a company would have in the big, fat Yellow Pages book. The salesman struggled with his pitch and clearly had little grasp of what the Internet actually was or how someone would find a business on it. The flimsy pitch got Musk thinking, and he reached out to Kimbal, talking up the idea of helping businesses get online for the first time.

“Elon said, ‘These guys don’t know what they are talking about. Maybe this is something we can do,’” Kimbal said. This was 1995, and the brothers were about to form Global Link Information Network, a start-up that would eventually be renamed Zip2. (For details on the controversy surrounding Zip2’s founding and Musk’s academic record, see Appendix 1.)

The Zip2 idea was ingenious. Few small businesses in 1995 understood the ramifications of the Internet. They had little idea how to get on it and didn’t really see the value in creating a website for their business or even in having a Yellow Pages–like listing online. Musk and his brother hoped to convince restaurants, clothing shops, hairdressers, and the like that the time had come for them to make their presence known to the Web-surfing public. Zip2 would create a searchable directory of businesses and tie this into maps. Musk often explained the concept through pizza, saying that everyone deserved the right to know the location of their closest pizza parlor and the turn-by-turn directions to get there. This may seem obvious today—think Yelp meets Google Maps—but back then, not even stoners had dreamed up such a service.

The Musk brothers brought Zip2 to life at 430 Sherman Avenue in Palo Alto. They rented a studio-apartment-sized office—twenty feet by thirty feet—and acquired some basic furniture. The three-story building had its quirks. There were no elevators, and the toilets often backed up. “It was literally a shitty place to work,” said an early employee. To get a fast Internet connection, Musk struck a deal with Ray Girouard, an entrepreneur who ran an Internet service provider operation from the floor below the Zip2 offices. According to Girouard, Musk drilled a hole in the drywall near the Zip2 door and then strung an Ethernet cable down the stairwell to the ISP. “They were slow to pay a couple of times but never stiffed me on the bill,” Girouard said.

Musk did all of the original coding behind the service himself, while the more amiable Kimbal looked to ramp up the door-to-door sales operation. Musk had acquired a cheap license to a database of business listings in the Bay Area that would give a business’s name and its address. He then contacted Navteq, a company that had spent hundreds of millions of dollars to create digital maps and directions that could be used in early GPS navigation-style devices, and struck a masterful bargain. “We called them up, and they gave us the technology for free,” said Kimbal. Musk merged the two databases together to get a rudimentary system up and running. Over time, Zip2’s engineers had to augment this initial data haul with more maps to cover areas outside of major metropolitan areas and to build custom turn-by-turn directions that would look good and work well on a home computer.

Errol Musk gave his sons $28,000 to help them through this period, but they were more or less broke after getting the office space, licensing software, and buying some equipment. For the first three months of Zip2’s life, Musk and his brother lived at the office. They had a small closet where they kept their clothes and would shower at the YMCA. “Sometimes we ate four meals a day at Jack in the Box,” Kimbal said. “It was open twenty-four hours, which suited our work schedule. I got a smoothie one time, and there was something in it. I just pulled it out and kept drinking. I haven’t been able to eat there since, but I can still recite their menu.”

Next, the brothers rented a two-bedroom apartment. They didn’t have the money or the inclination to get furniture. So there were just a couple of mattresses on the floor. Musk somehow managed to convince a young South Korean engineer to come work at Zip2 as an intern in exchange for room and board. “This poor kid thought he was coming over for a job at a big company,” Kimbal said. “He ended up living with us and had no idea what he was getting into.” One day, the intern drove the Musks’ battered BMW 320i to work, and a wheel came off en route. The axle dug into the street at the intersection of Page Mill Road and El Camino Real, and the groove it carved out remained visible for years.

Zip2 may have been a go-go Internet enterprise aimed at the Information Age, but getting it off the ground required old-fashioned door-to-door salesmanship. Businesses needed to be persuaded of the Web’s benefits and charmed into paying for the unknown. In late 1995, the Musk brothers began making their first hires and assembling a motley sales team. Jeff Heilman, a free-spirited twenty-year-old trying to figure out what to do with his life, arrived as one of Zip2’s first recruits. He’d been watching TV late one night with his dad and seen a Web address printed at the bottom of the screen during a commercial. “It was for something dot-com,” Heilman said. “I remember sitting there and asking my dad what we were looking at. He said he didn’t know, either. That’s when I realized I had to go find me some Internet.” Heilman spent a couple of weeks trying to chat up people who could explain the Internet to him and then stumbled on a two-by-two-inch Zip2 job listing in the San Jose Mercury News. “Internet Sales Apply Here!” it read, and Heilman got the gig. A handful of other salespeople joined him and worked for commissions.

Musk never seemed to leave the office. He slept, not unlike a dog, on a beanbag next to his desk. “Almost every day, I’d come in at seven thirty or eight A.M., and he’d be asleep right there on that bag,” Heilman said. “Maybe he showered on the weekends. I don’t know.” Musk asked those first employees of Zip2 to give him a kick when they arrived, and he’d wake up and get back to work. While Musk did his possessed coder thing, Kimbal became the rah-rah sales leader. “Kimbal was the eternal optimist, and he was very, very uplifting,” Heilman said. “I had never met anyone quite like him.” Kimbal sent Heilman to the high-end Stanford shopping mall and to University Avenue, the main drag in Palo Alto, to coax retailers into signing up with Zip2, explaining that a sponsored listing would send a company to the top of search results. The big problem, of course, was that no one was buying. Week after week, Heilman knocked on doors and returned to the office with very little to report in the way of good news. The nicest responses came from the people who told Heilman that advertising on the Internet sounded like the dumbest thing they had ever heard of. Most often, the shop owners just told Heilman to leave and stop bothering them. When lunchtime came around, the Musks would reach into a cigar box where they kept some cash, take Heilman out, and get the depressing status reports on the sales.

Craig Mohr, another early employee, gave up his job selling real estate to hawk Zip2’s service. He decided to court auto dealerships because they usually spent lots of money on advertising. He told them about Zip2’s main website—www.totalinfo.com—and tried to convince them that demand was high to get a listing like www.totalinfo.com/toyotaofsiliconvalley. The service did not always work when Mohr demonstrated it or it would load very slowly, as was common back then. This forced him to talk the customers into imagining Zip2’s potential. “One day I came back with about nine hundred dollars in checks,” Mohr said. “I walked into the office and asked the guys what they wanted me to do with the money. Elon stopped pounding his keyboard, leaned out from behind his monitor, and said, ‘No way, you’ve got money.’”

What kept the employees’ spirits up were the continuous improvements Musk made with the Zip2 software. The service had morphed from a proof-of-concept to an actual product that could be used and demoed. Ever marketing savvy, the Musk brothers tried to make their Web service seem more important by giving it an imposing physical body. Musk built a huge case around a standard PC and lugged the unit onto a base with wheels. When prospective investors would come by, Musk would put on a show and roll this massive machine out so that it appeared like Zip2 ran inside of a mini-supercomputer. “The investors thought that was impressive,” Kimbal said. Heilman also noticed that the investors bought into Musk’s slavish devotion to the company. “Even then, as essentially a college kid with zits, Elon had this drive that this thing—whatever it was—had to get done and that if he didn’t do it, he’d miss his shot,” Heilman said. “I think that’s what the VCs saw—that he was willing to stake his existence on building out this platform.” Musk actually said as much to one venture capitalist, informing him, “My mentality is that of a samurai. I would rather commit seppuku than fail.”

Early on in the Zip2 venture, Musk acquired an important confidant, who tempered some of these more dramatic impulses. Greg Kouri, a Canadian businessman in his mid-thirties, had met the Musks in Toronto and bought into the early Zip2 brainstorming. The boys had showed up at his door one morning to inform Kouri that they intended to head to California to give the business a shot. Still in his red bathrobe, Kouri went back into the house, dug around for a couple of minutes, and came back with a wad of $6,000. In early 1996, he moved to California and joined Zip2 as a cofounder.

Kouri, who had done a number of real estate deals in the past and had actual business experience and skills at reading people, served as the adult supervision at Zip2. The Canadian had a knack for calming Musk and ended up becoming something of a mentor. “Really smart people sometimes don’t understand that not everyone can keep up with them or go as fast,” said Derek Proudian, a venture capitalist who would become Zip2’s chief executive officer. “Greg is one of the few people that Elon would listen to and had a way of putting things in context for him.” Kouri also used to referee fistfights between Elon and Kimbal, in the middle of the office.

“I don’t get in fights with anyone else, but Elon and I don’t have the ability to reconcile a vision other than our own,” Kimbal said. During a particularly nasty scrap over a business decision, Elon ripped some skin off his fist and had to go get a tetanus shot. Kouri put an end to the fights after that. (Kouri died of a heart attack in 2012 at the age of fifty-one, having made a fortune investing in Musk’s companies. Musk attended his funeral. “We owe him a lot,” said Kimbal.)

In early 1996, Zip2 underwent a massive change. The venture capital firm Mohr Davidow Ventures had caught wind of a couple of South African boys trying to make a Yellow Pages for the Internet and met with the brothers. Musk, while raw in his presentation skills, pitched the company well enough, and the investors came away impressed with his energy. Mohr Davidow invested $3 million into the company.* With these funds in hand, the company officially changed its name from Global Link to Zip2—the idea being zip to here, zip to there—moved to a larger office at 390 Cambridge Avenue in Palo Alto, and began hiring talented engineers. Zip2 also shifted its business strategy. At the time, the company had built one of the best direction systems on the Web. Zip2 would advance this technology and take it from focusing just on the Bay Area to having a national scope. The company’s main focus, however, would be an altogether new play. Instead of selling its service door-to-door, Zip2 would create a software package that could be sold to newspapers, which would in turn build their own directories for real estate, auto dealers, and classifieds. The newspapers were late understanding how the Internet would impact their businesses, and Zip2’s software would give them a quick way of getting online without needing to develop all their own technology from scratch. For its part, Zip2 could chase bigger prey and get a cut of a nationwide network of listings.

This transition of the business model and the company’s makeup would be a seminal moment in Musk’s life. The venture capitalists pushed Musk into the role of chief technology officer and hired Rich Sorkin as the company’s CEO. Sorkin had worked at Creative Labs, a maker of audio equipment, and run the business development group at the company, where he steered a number of investments in Internet start-ups. Zip2’s investors saw him as experienced and clued in to the Web. While Musk agreed to the arrangement, he came to resent giving up control of Zip2. “Probably the biggest regret the whole time I worked with him was that he had made a deal with the devil with Mohr Davidow,” said Jim Ambras, the vice president of engineering at Zip2. “Elon didn’t have any operational responsibilities, and he wanted to be CEO.”

Ambras had worked at Hewlett-Packard Labs and Silicon Graphics Inc. and exemplified the high-caliber talent Zip2 brought on after the first wave of money arrived. Silicon Graphics, a maker of high-end computers beloved by Hollywood, was the flashiest company of its day and had hoarded the elite geeks of Silicon Valley. And yet Ambras used the promise of Internet riches to poach a team of SGI’s smartest engineers over to Zip2. “Our attorneys got a letter from SGI saying that we were cherry-picking the very best guys,” Ambras said. “Elon thought that was fantastic.”

While Musk had exceled as a self-taught coder, his skills weren’t nearly as polished as those of the new hires. They took one look at Zip2’s code and began rewriting the vast majority of the software. Musk bristled at some of their changes, but the computer scientists needed just a fraction of the lines of code that Musk used to get their jobs done. They had a knack for dividing software projects into chunks that could be altered and refined whereas Musk fell into the classic self-taught coder trap of writing what developers call hairballs—big, monolithic hunks of code that could go berserk for mysterious reasons. The engineers also brought a more refined working structure and realistic deadlines to the engineering group. This was a welcome change from Musk’s approach, which had been to set overly optimistic deadlines and then try to get engineers to work nonstop for days on end to meet the goals. “If you asked Elon how long it would take to do something, there was never anything in his mind that would take more than an hour,” Ambras said. “We came to interpret an hour as really taking a day or two and if Elon ever did say something would take a day, we allowed for a week or two weeks.”

Starting Zip2 and watching it grow imbued Musk with self-confidence. Terence Beney, one of Musk’s high school friends, came to California for a visit and noticed the change in Musk’s character right away. He watched Musk confront a nasty landlord who had been giving his mother, who was renting an apartment in town, a hard time. “He said, ‘If you’re going to bully someone, bully me.’ It was startling to see him take over the situation. The last time I had seen him he was this geeky, awkward kid who would sometimes lose his temper. He was the kid you would pick on to get a response. Now he was confident and in control.” Musk also began consciously trying to manage his criticism of others. “Elon is not someone who would say, ‘I feel you. I see your point of view,’” said Justine. “Because he doesn’t have that ‘I feel you’ dimension there were things that seemed obvious to other people that weren’t that obvious to him. He had to learn that a twenty-something-year-old shouldn’t really shoot down the plans of older, senior people and point out everything wrong with them. He learned to modify his behavior in certain ways. I just think he comes at the world through strategy and intellect.” The personality tweaks worked with varying degrees of success. Musk still tended to drive the young engineers mad with his work demands and blunt criticism. “I remember being in a meeting once brainstorming about a new product—a new-car site,” said Doris Downes, the creative director at Zip2. “Someone complained about a technical change that we wanted being impossible. Elon turned and said, ‘I don’t really give a damn what you think,’ and walked out of the meeting. For Elon, the word no does not exist, and he expects that attitude from everyone around him.” Periodically, Musk let loose on the more senior executives as well. “You would see people come out of the meetings with this disgusted look on their face,” Mohr, the salesman, said. “You don’t get to where Elon is now by always being a nice guy, and he was just so driven and sure of himself.”

As Musk tried to come to terms with the changes the investors had inflicted on Zip2, he did enjoy some of the perks of having big-money backing. The financiers helped the Musk brothers with their visas. They also gave them $30,000 each to buy cars. Musk and Kimbal had traded in their dilapidated BMW for a dilapidated sedan that they spray-painted with polka dots. Kimbal upgraded from that to a BMW 3 Series, and Musk bought a Jaguar E-Type. “It kept breaking down, and would arrive at the office on a flatbed,” Kimbal said. “But Elon always thought big.”*

As a bonding exercise one weekend, Musk, Ambras, a few other employees and friends took off for a bike ride through the Saratoga Gap trail in the Santa Cruz Mountains. Most of the riders had been training and were accustomed to strenuous sessions and the summer’s heat. They set up the mountains at a furious pace. After an hour, Russ Rive, Musk’s cousin, reached the top and proceeded to vomit. Right behind him were the rest of the cyclists. Then, fifteen minutes later, Musk became visible to the group. His face had turned purple, and sweat poured out of him, and he made it to the top. “I always think back to that ride. He wasn’t close to being in the condition needed for it,” Ambras said. “Anyone else would have quit or walked up their bike. As I watched him climb that final hundred feet with suffering all over his face, I thought, That’s Elon. Do or die but don’t give up.”

Musk continued to be a ball of energy around the office as well. Ahead of visits by venture capitalists and other investors, Musk would rally the troops and instruct them all to get on the phone to create a buzzy atmosphere. He also formed a video-game team to participate in competitions around Quake, a first-person-shooter game. “We competed in one of the first nationwide tournaments,” Musk said. “We came in second, and we would have come in first, but one of our top players’ machine crashed because he had pushed his graphics card too hard. We won a few thousand dollars.”

Zip2 had remarkable success courting newspapers. The New York Times, Knight Ridder, Hearst Corporation, and other media properties signed up to its service. Some of these companies contributed $50 million in additional funding for Zip2. Services like Craigslist with its free online classifieds had just started to appear, and the newspapers needed some course of action. “The newspapers knew they were in trouble with the Internet, and the idea was to sign up as many of them as possible,” Ambras said. “They wanted classifieds and listings for real estate, automotive, and entertainment and could use us as a platform for all these online services.” Zip2 acquired a trademark for its “We Power the Press” slogan and the influx of cash kept Zip2 growing fast. Company headquarters were soon so crowded that one desk ended up directly in front of the women’s bathroom. In 1997, Zip2 moved into flashier, more spacious digs at 444 Castro Street in Mountain View.

It irritated Musk that Zip2 had become a behind-the-scenes player to the newspapers. He believed the company could offer interesting services directly to consumers and encouraged the purchase of the domain name city.com with the hopes of turning it into a consumer destination. But the lure of the media companies’ money kept Sorkin and the board on a conservative path, and they decided to worry about a consumer push down the road.

In April 1998, Zip2 announced a blockbuster move to double down on its strategy. It would merge with its main competitor CitySearch in a deal valued at around $300 million. The new company would retain the CitySearch name, while Sorkin would head up the venture. On paper, the union looked very much like a merger of equals. CitySearch had built up an extensive set of directories for cities around the country. It also appeared to have strong sales and marketing teams that would complement the talented engineers at Zip2. The merger had been announced in the press and seemed inevitable.

The opinions on what happened next vary greatly. The logistics of the situation required the two companies to go over each other’s books and to figure out which employees would be fired to avoid a duplication of roles. This process raised some questions about how frank CitySearch had been with its financials and rankled some executives at Zip2 who could see their positions being diminished or erased altogether at the new company. One faction inside Zip2 argued that the deal should be abandoned, while Sorkin demanded that it go through. Musk, who had been an early advocate of the deal, turned against it. In May 1998, the two companies canceled the merger, and the press pounced, making a big deal of the chaotic bust-up. Musk urged Zip2’s board to oust Sorkin and reinstate him as CEO of Zip2. The board declined. Instead, Musk lost his chairman title, and Sorkin was replaced by Derek Proudian, a venture capitalist with Mohr Davidow. Sorkin considered Musk’s behavior through the whole affair atrocious and later pointed to the board’s reaction and Musk’s demotion as evidence that they felt the same way. “There was a lot of backlash and finger-pointing,” Proudian said. “Elon wanted to be CEO, but I said, ‘This is your first company. Let’s find an acquirer and make some money, so you can do your second, third, and fourth company.’”

With the deal busted, Zip2 found itself in a predicament. It was losing money. Musk still wanted to go the consumer route, but Proudian feared that would take too much capital. Microsoft had mounted a charge into the same market, and start-ups with mapping, real estate, and automotive ideas multiplied. The Zip2 engineers were deflated and worried that they might not be able to outrun the competition. Then, in February 1999, the PC maker Compaq Computer suddenly offered to pay $307 million in cash for Zip2. “It was like pennies from heaven,” said Ed Ho, a former Zip2 executive. Zip2’s board accepted the offer, and the company rented out a restaurant in Palo Alto and threw a huge party. Mohr Davidow had made back twenty times its original investment, and Musk and Kimbal had come away with $22 million and $15 million, respectively. Musk never entertained the idea of sticking around at Compaq. “As soon as it was clear the company would be sold, Elon was on to his next project,” Proudian said. From that point on, Musk would fight to maintain control of his companies and stay CEO. “We were overwhelmed and just thought these guys must know what they’re doing,” Kimbal said. “But they’ didn’t. There was no vision once they took over. They were investors, and we got on well with them, but the vision had just disappeared from the company.”

Years later, after he had time to reflect on the Zip2 situation, Musk realized that he could have handled some of the situations with employees better. “I had never really run a team of any sort before,” Musk said. “I’d never been a sports captain or a captain of anything or managed a single person. I had to think, Okay, what are the things that affect how a team functions. The first obvious assumption would be that other people will behave like you. But that’s not true. Even if they would like to behave like you, they don’t necessarily have all the assumptions or information that you have in your mind. So, if I know a certain set of things, and I talk to a replica of myself but only communicate half the information, you can’t expect that the replica would come to the same conclusion. You have to put yourself in a position where you say, ‘Well, how would this sound to them, knowing what they know?’”

Employees at Zip2 would go home at night, come back, and find that Musk had changed their work without talking to them, and Musk’s confrontational style did more harm than good. “Yeah, we had some very good software engineers at Zip2, but I mean, I could code way better than them. And I’d just go in and fix their fucking code,” Musk said. “I would be frustrated waiting for their stuff, so I’m going to go and fix your code and now it runs five times faster, you idiot. There was one guy who wrote a quantum mechanics equation, a quantum probability on the board, and he got it wrong. I’m like, ‘How can you write that?’ Then I corrected it for him. He hated me after that. Eventually, I realized, Okay, I might have fixed that thing but now I’ve made the person unproductive. It just wasn’t a good way to go about things.”

Musk, the dot-com striver, had been both lucky and good. He had a decent idea, turned it into a real service, and came out of the dot-com tumult with cash in his pockets, which was better than what many of his compatriots could say. The process had been painful. Musk had yearned to be a leader, but the people around him struggled to see how Musk as the CEO could work. As far as Musk was concerned, they were all wrong, and he set out to prove his point with what would end up being even more dramatic results.