The New Space Race: One Man's Mission to Build a Galactic Internet

Google, Facebook, and Elon Musk want to wire the world. A guy in Florida may beat them to it

Agefoto(2); NASA(3); Science Source(3)

 

 

Greg Wyler is friendly and gregarious and talks about his business plans with a fervor that borders on the religious. He’s also frenetic and hard to pin down. So when the technology entrepreneur, who’s based in Sewall’s Point, Fla., suggested our interview take place in Puerto Rico, I accepted. He and his family had chartered a jet and planned to stop in San Juan on their way to St. Bart’s. While his wife and kids had lunch on the plane, we’d chat in the airport about the ambitious satellite company he’s starting. “Bring a cheese pizza and a sausage one,” he texted. “The kids will love you!” Ten minutes later: “And a Diet Coke.”

For weeks, people who knew Wyler kept telling me he’s an out-there, creative thinker who can come across as disorganized and impulsive. Roll with it, they’d say, because he’s a genius. So it wasn’t a total surprise when he texted again to cancel—they’d departed later than expected—and I found myself alone in a San Juan hotel with two pizzas and a Diet Coke. “So sorry, tried,” he wrote. “It’s been a busy morning …”

For a guy who can’t seem to keep a schedule, Wyler, 45, has had a remarkable run launching a series of companies, each more daring than the last. After making a fortune designing and selling computer parts in his 20s, he decided to dedicate his life to connecting the unconnected. He founded a telecommunications startup that laid fiber-optic cables throughout Rwanda, then a satellite company to bring high-speed, affordable Internet service to islands and other remote locales. Now he’s starting a third venture, OneWeb, with engineering headquarters in Silicon Valley, that aspires to create an elaborate array of low-orbit satellites to bring Internet access to everyone on earth.

In tech-conference speak, what Wyler’s been doing for the past dozen years is connecting “the other 3 billion.” This is the half of the world’s population that for various reasons is not online. Companies such as Google and Facebook have received a lot of attention for their philanthropic plans to bring the Internet to the developing world. Mark Zuckerberg, for example, invited Time to follow him around rural India for a story about his evolution as a philanthropist and business leader, though he provided few specifics on what Facebook intends to do for India’s poor. And Elon Musk—he of Tesla Motors, SpaceX, and the Hyperloop—has received a wave of press, and $1 billion in capital from Google and Fidelity—after unveiling his own space Internet plan.

Wyler is more of a fringe character, and he’s been preaching since well before it was cool that Internet access ought to be a human right. He’s probably the only person alive who’s done the manual labor on the ground and the big, architectural thinking up above to know just how messy wiring the entire planet will get. And with OneWeb, Wyler may receive his due. It’s hard to ignore an Internet delivered from the heavens.

“People handed me the Zuckerberg article, and other people called me after it came out,” he says after we finally link up. “This is great. He has a much bigger pedestal than I do. I’ve been trying to get people to understand that connectivity is a fundamental layer for societal and economic growth.” Then, being as polite as possible, he adds: “The other thing that’s great is that I know our system works.”

Wyler sold his first company for $100 million.
Wyler sold his first company for $100 million.
Photographer: Stian Rasmussen for Bloomberg Businessweek

Wyler smiles a lot. That, combined with his exaggerated features—out-there chin, puffy cheeks—make him look like a caricature by a sidewalk artist. He talks quickly and incessantly and prefers to walk while talking. He grew up near Boston, the son of a prominent, aggressive insurance litigator and a doting mom. When he felt like it, Wyler did well in class, but his mind often wandered, and his grades fluctuated between A’s and “complete averageness.” He’s capable of extreme focus if a subject interests him, which is what happened in high school when he got into computers and cars. He taught himself how to use computer-aided design (CAD) software, which had a multiplier effect on his innate ability to imagine complex devices. Soon enough, he was designing things. One of his first ideas was a radical fuel-injection system for a Ford Mustang. He found the name of Ford’s head of engineering in a technical journal and called him out of the blue. “We talked cars, and it was this eye-opening experience that, OK, there is this whole other world out there,” Wyler says. “Pre-Internet it was hard to find people that were as passionately and deeply interested in arcane topics.”

After high school, Wyler bounced from Xavier University in Cincinnati to North Adams State College in western Massachusetts. He studied finance and computer science and then went to law school at the Illinois Institute of Technology in Chicago. It was there, in 1992, that he hit on his first really big business idea: a new take on the heat sink, the hunk of metal that draws heat away from a PC’s main processor. He started a company, called No Overhead Computers, and built custom PCs while refining his heat-sink design. He eventually developed a PC that could run without needing one of the big, noisy fans that were ubiquitous in computers of the time.

In 1994, Wyler created a company called Silent Systems to sell his heat sink. “He didn’t call me once a day, he called me three to five times a day,” says Jim Rappaport, the director of the New Boston Fund, whom Wyler pursued as a potential investor. “He’s indefatigable.” Rappaport relented, and Wyler wowed him with a demonstration of the Silent Systems machine vs. an IBM PC. They went into business together. Wyler commuted between Chicago and Boston, finishing law school while he built up the company. He often worked 36 hours straight, fashioning a kind of cubicle cave that blocked out all light so he could concentrate on his computer screen. After five years, Silent Systems managed to create the best heat sink in the PC industry, at a cost of 50¢ per unit, about 80 percent cheaper than its rivals. Silent Systems later secured Dell, Hewlett-Packard, and others as customers and was eventually acquired by electronics maker Molex for about $100 million. Wyler, still in his mid-20s, was a multimillionaire. Flush with cash, he dabbled in real estate and hit it big with a couple of quick-strike tech investments during the dot-com boom. Then everything changed.

On Oct. 1, 2002, Wyler drove to his mother’s house in Winchester, Mass. Susann hadn’t been answering his phone calls, and he felt something was wrong. He arrived to find the front door open and the house silent. He made his way through the house and into the garage, where he found his mother’s body soaked in blood and her head crushed. Someone had bludgeoned her to death.

For Wyler, there was no mystery in what had happened. When the police arrived, he blurted out to a detective, “He did it. My father finally killed my mother,” according to an exposé on the crime in Boston Magazine. In the article, Wyler described his childhood battles with his father, Geoffrey, and leaving home as a teenager to live with his grandparents. He alleged that his father had a history of physically and psychologically abusing his mother, that his parents’ relationship had severely worsened in the year before the murder as Susann filed for divorce, and that Geoffrey had recently decided to sue Greg for a piece of the Silent Systems windfall—all possible reasons for a confrontation. Geoffrey Wyler has denied the accusations that he abused Susann.

“Someone went into my mother’s house, beat her to death, and left without taking a single thing,” Wyler told the magazine. “It all points in one direction, and one direction only.” The police never accused his father of any crime, and to this day, no one has been charged. Geoffrey Wyler declined to comment.

Today, Greg Wyler avoids talking about the murder. When it comes up, he says only that the event helped set his life on a different course. “My mom’s death played a role in my thinking about doing something more, something greater,” he says. “I decided that whatever I did next had to have a mission.”

In late 2002, while attending a friend’s wedding in Boston, Wyler met Theogene Rudasingwa, then the chief of staff to Rwandan President Paul Kagame. The two hit it off, and Wyler sensed that he’d found his mission: Help a post-genocide Rwanda modernize. He began thinking up ways to connect more of its schools to the Internet, then he started dreaming about turning the country into a technology hub. “The mindset in the world at the time was that Internet infrastructure was not a high priority,” Wyler says. “I thought that was wrong. When you have good Internet access, you have economic growth.”

About a year after his mother’s murder and the chance wedding encounter, Wyler founded Terracom, a telecommunications company whose business plan was to bring cell phone and Internet service to Rwandans. He managed the company from the U.S. but made frequent trips to Africa. Workers—including Wyler, on occasion—dug trenches to lay hundreds of miles of fiber-optic cable and set up Africa’s first 3G cellular network. Terracom’s service ran faster and cost less than the government-backed RwandaTel and, after a year in business, had more subscribers. In 2005, Terracom acquired RwandaTel for $20 million.

For a while, President Kagame lauded Wyler’s work as the centerpiece of a broad modernization plan. But the merger of the two companies became a management nightmare. “RwandaTel was hugely overstaffed, had horrible equipment, and was a total disaster,” says John Dick, a major investor in Terracom and board member of Liberty Global, a London-based cable company with $20 billion in annual revenue. Wyler and Dick asked the Chinese telecommunications-equipment maker Huawei Technologies to help design a more efficient, more economical system, but the venture continued to struggle, and some members of the Rwandan government grumbled about Wyler managing the business from afar. “It was obvious that the resulting company needed to be owned by Rwanda,” Dick says. In 2007 the Rwandan government bought the merged company. “I think we made a lot of progress, but there were some political aspects that became hard to manage,” Wyler says. “It became easier to sell the assets, and by then I wasn’t even involved.”

Wyler says he checked out not because of disinterest but because of a technical limitation Terracom couldn’t solve. The biggest bottleneck of Terracom’s fiber network was where it connected with the global Internet. Data could zip around Rwanda, but getting it in or out of the country required slow, expensive satellite connections. Wyler figured that many countries must be in a similar predicament, and his mission expanded. He decided to start a company that would go beyond bringing just one country online—he’d connect countries to the rest of the world.

New Space Race: Mission to Build Galactic Internet

It takes an hour to drive from the Félix Eboué Airport in Cayenne, French Guiana, to the coastal town of Kourou. There isn’t much to see along the way, mostly Amazonian rain forest with its million shades of green. Every now and then, a billboard appears that advertises rockets or rocket parts from Arianespace, Airbus, or some other European aerospace company. In Kourou, you take a left to enter the Guiana Space Centre, one of the busiest spaceports in the world.

The only reason to build a spaceport in a jungle is physics. Kourou is about 300 miles north of the equator, which lessens the fuel needed for a rocket to place an object in an equatorial orbit, where the earth’s rotational speed is fastest and distance to orbit is shortest. Because of the savings, a rocket launched from Kourou can carry payloads with 20 percent to 35 percent more mass than the same rocket launched from other major spaceports.

European nations worked together to build the facility, which occupies 270 square miles along the Atlantic Ocean. This location provides the added bonus that if a rocket malfunctions, there’s plenty of room to blow it up over the sea. The spaceport has multiple launchpads, massive assembly buildings, liquid oxygen and kerosene production centers, and mission control rooms. Since hunting is not allowed in the area, there are also plenty of things a rocket company doesn’t need, such as sloths and monkeys in the trees and fat snakes lazing by the side of the road.

A Soyuz rocket carrying O3b satellites lifts off in French Guiana.
A Soyuz rocket carrying O3b satellites lifts off in French Guiana.
Source: ESA-CNES-Arianespace/Optique Video Du CSG/P. Baudon

On Dec. 18 hundreds of locals file into the main mission control center to watch the launch of a Russian-made Soyuz rocket. Launch days in French Guiana are a principal form of entertainment, and women in high heels and sundresses, along with military men in uniform, pack the theater area of mission control to watch and listen to the play-by-play from an announcer at the front of the room. The Soyuz’s payload is four satellites belonging to O3b, a startup founded by Wyler in 2007. It’s the company’s third launch; the plan is to position the satellites with eight others already in orbit in a ring around the equator.

It’s long been possible to get the Internet via satellite, with service about as speedy and responsive as an early-1990s modem. Data are transmitted back and forth between terrestrial antennas and satellites in geosynchronous orbits about 22,000 miles up. At that distance, it takes more than 500 milliseconds for the signal to get to a satellite and back. While a half-second might not sound like much, it’s enough to make Skype, FaceTime, games, and any modern cloud-based application torturous.

Wyler’s gamble is to place O3b’s satellites in what’s known as a medium-earth orbit, which is about 5,000 miles into space. At that distance, data can travel up and back in about 150 milliseconds, a performance comparable to fiber-optic cable. There’s a trade-off: The lower the satellite, the less earth it can see. So O3b is putting up an unusually large number of satellites, 12 so far, with more to come. The cost to launch four satellites—each the size of a large restaurant-grade refrigerator—runs about $300 million, and the company has raised more than $1.3 billion to date from the likes of Google, HSBC, and SES, one of the world’s largest satellite operators.

http://www.bloomberg.com/graphics/2015-satellite-network/embed/o3b.html?v=3

The December launch went off with only a couple of minor hitches. For about 30 minutes, the Russian team leader had a phone to each ear, and O3b Chief Executive Officer Steve Collar paced around the theater because the satellites went quiet. But the machines ultimately came to life as planned, and O3b’s full complement of satellites started sending down 120 Internet beams that the company can direct at will. Each beam creates a coverage area of 400 miles in diameter: Anyone within that circle can get the Internet at fiber speed. Typically, a telecommunications company will sign a deal with O3b for a beam, put up a 14-foot antenna to receive the signal, and then distribute the bandwidth to its business and consumer customers via cellular towers and cables.

The arrival of O3b’s service has been a boon for islands throughout the Pacific. The Cook Islands, for example, used to rely on geostationary satellite connections. Movies wouldn’t stream, online games didn’t work, and when a hospital tried to have a video call with a specialist in New Zealand, the picture was delayed and pixelated. “The service is reliable, but it’s slow and very expensive,” says Jules Maher, the former CEO of Telecom Cook Islands, the sole telecommunications provider. Maher looked into an undersea fiber cable to connect the main island to Tahiti, but the price was crippling. “There are only 10,000 residents on the main island, and the cost would be $30 million, so it was hard to see how we would ever get a return on the investment.”

 

Maher’s company has been an O3b customer since March. He’s impressed. “Businesses are paying 10 percent of what they used to pay and getting 12 times the speed,” he says. The service is already having an effect on demographics—some of the younger Cook Islanders are opting to stay at home rather than go to New Zealand or Australia for school, in part because they can now take classes online. They’re losing the sense that, if they stay, the world will pass them by. “No one wants to feel like they’re in a backwater,” says Maher. Telcos in Papua New Guinea, the Democratic Republic of the Congo, American Samoa, Malaysia, and Afghanistan are among the 35 customers O3b has signed up.

O3b is also finding eager customers among cruise lines and offshore oil rigs. Cruise ships typically charge 75¢ to $1 per minute for geostationary Internet access. “You have to take out a loan to stay on the Internet, and I am only partially kidding,” says Bill Martin, the chief information officer for Royal Caribbean Cruises. Royal Caribbean now has beams from O3b following three of its ships and is giving passengers free access as an enticement over rival cruise lines. Martin rattles off anecdotes of people streaming a movie while on the treadmill, catching a college football game while grabbing a drink, and Snapchatting while onboard—they now have all their usual obsessions and time wasters at sea. “The younger generation does not want to be disconnected,” he says. “This is an incredible edge for us.” The amount of bandwidth going to just one Royal Caribbean ship surpasses that reaching all the rest of the cruise ships on the ocean combined.

With the 12 satellites up and running, O3b has moved from doing trials throughout 2014 to launching its full commercial service in December. Prices for the service fluctuate depending on what its telco customers charge their subscribers, but most often consumers end up paying about the same price as they would for a fiber connection. In a matter of months, O3b has emerged as the largest Internet service provider in the Pacific, and it should break even by the middle of the year on $100 million in annual revenue, according to the company. “We can launch more than 100 satellites into the constellation,” says CEO Collar. “And as we launch more satellites, our network becomes more efficient, the price goes down.”

http://www.bloomberg.com/graphics/2015-satellite-network/embed/oneweb.html?v=3

A couple of weeks after the O3b launch, Wyler is saying—with a straight face—how exhilarating it feels to be doing a startup in a garage. The garage in question is attached to his West Coast home, a mansion in Atherton, Calif., the toniest of all the Silicon Valley towns. The whole first floor of the spread is the temporary headquarters of OneWeb, his newest company.

There’s a handful of mechanical engineers in the living room, some communications experts in the dining room, and a couple of satellite specialists at the kitchen table. Ethernet cables run all over the floor. In the garage, there’s a workshop with a 3D printer and what looks like a 15-foot-long barbecue spit. Engineers use it to assemble what will become a 280-pound satellite. For break time, there’s a putting green and a boccie ball court on the property. (The company’s legal name is WorldVu, although it will operate under the OneWeb brand.)

OneWeb is a supercharged version of O3b. Instead of dozens of satellites, Wyler plans to put up hundreds—648, to start with. The satellites will be in a low-earth orbit 750 miles up, much closer than even O3b’s machines. Engineers expect data to travel between space and the surface in 20 milliseconds, which would provide a state-of-the-art Internet service capable of handling any application. While sitting by the fireplace in his guesthouse, Wyler holds a gray, semispherical object about the size of a car tire. It’s a OneWeb rooftop antenna. Unlike a typical satellite TV dish, which requires an installer to aim the dish carefully to get reception, Wyler says, his device just has to be put on the roof, round side up. OneWeb will have so many satellites, he explains, you’d have a hard time not getting a signal. So where O3b requires large, specialized antennas run by telcos, OneWeb will offer this simple device for use by individuals and public and commercial buildings, such as schools, stores, and hospitals. It will act as a kind of local Internet hub connecting with nearby devices via Wi-Fi or a cellular signal.

A payload of four O3b satellites getting prepped for launch.
A payload of four O3b satellites getting prepped for launch.
Photograph: Courtesy ESA-CNES-Arianespace/Optique Video du CSG/P. Baudon

“You don’t have to buy an antenna,” Wyler says. “You just have to be near a school or a health center, and your phone or tablet will log on.” He expects the antennas to start at about $200 each and promises they’ll be durable and easy to use. “This thing can sit in the mud for months on end, it’s waterproof, and it has no buttons,” he says. “You should not need any words to figure out how to turn this on.”

There will certainly be some overlap between OneWeb and O3b, but Wyler sees them as complementary services that cater to different markets. O3b will be more business-to-business, offering large amounts of bandwidth to countries, telcos, and large ships. If a ship is within range, O3b can give it capacity that would be difficult for OneWeb to match. OneWeb, though, will have much broader coverage and serve both business customers and consumers. Wyler remains a large shareholder in O3b.

On another level, OneWeb could function as a global Internet backup system. If a bunch of fiber cables get cut and a region loses its Internet connection, OneWeb can pick up the traffic. The network should also deliver much faster Internet service to airplanes, and it would be of great use in a natural disaster when terrestrial communication systems are suddenly wiped out. OneWeb could theoretically drop off dozens of its antennas, point them skyward, and establish instant Internet for emergency workers and others.

Wyler says he’s not trying to compete as a global telecommunications company. He’d prefer to stay in the wholesale end of the business, selling antennas and satellite service to existing telcos around the world, who would then resell the antennas and the Internet service that comes with them. The telcos, and not OneWeb, will set the price on the service. “We hope to see prices that are affordable to the consumer,” he says.

A lot can go wrong before OneWeb’s network is complete. A couple of companies—Teledesic and SkyBridge—tried 10-plus years ago to build similar networks and burned through billions of dollars before failing and scaring investors away from the idea for years. Wyler and others argue that these efforts were ahead of their time, and that the underlying technology has improved enough to make the idea, with a couple of architectural tweaks, feasible again. OneWeb says it will be able to cover an area the size of India with three satellites. The machines, though, are always on the move in a gridlike pattern. As such, the venture must develop ways to pass a communication signal from one satellite to the next and run millions of calculations every minute to figure out how to best divvy up bandwidth among all the people tapping into one satellite.

Only a handful of companies build satellites, and they’re usually one-offs designed for a specific purpose. OneWeb will need one of these manufacturers to produce its machines at scale. It will also need to work with most of the major rocket-launch companies to meet its unprecedented goal of sending up a new satellite every 20 days. “This is the biggest thing that has ever been done in the satellite industry,” says David Bettinger, who left his job as chief technology officer at IDirect, a satellite communications company, to join Wyler. “It takes a Greg to do something like this.”

Wyler, who’s put in $6 million of his own money so far, expects it will take more than $2 billion to get OneWeb going. The company has lined up Virgin Group and Qualcomm as investors, with each putting in “tens of millions,” according to Virgin founder Richard Branson, who’s joined the OneWeb board. “We have the capacity to put up nearly 2,500 satellites,” Branson says. “If we have our figures right, this will be a highly profitable business that also encompasses charity and delivers a much-needed service.”

Wyler hopes to have OneWeb up and running by 2018. In the meantime, rivals will be expanding their efforts to connect the rest of the world. Google had at one point looked to fund OneWeb and make it part of the company’s broad Internet connection efforts, but Wyler and Google CEO Larry Page decided to part ways, Branson says. Google is charging ahead with its own plan, Project Loon, in which giant weather balloons rigged with communications gear will float above remote areas to create a wireless network. Facebook has a number of schemes, too, and has been exploring everything from drones and lasers to more cleverly placed cell towers in rural areas to bridge gaps in Internet connectivity.

The biggest challenge may come from Musk, who used to crash in Wyler’s guesthouse and has just announced plans to create his own version of an Internet space network. Musk’s plan is to build thousands of satellites at a SpaceX factory, launch them with his own rockets, and use them to handle much of the world’s Internet traffic. “We want a satellite that is an order of magnitude more sophisticated than what Greg wants,” he says. “I think there should be two competing systems.”

Branson counters that Wyler’s the only person to have thought through all of the technical issues and acquired the international wireless spectrum rights to provide Internet service from space. “I don’t think Elon can do a competing thing,” says Branson, who’s friends with both men. “If Elon wants to get into this area, the logical thing for him would be to tie up with us.”

During a walk-and-talk interview with Wyler in Atherton, the conversation turns to his reputation as brilliant but flighty. “I am a shepherd,” he says, explaining that he’s good at figuring how things will work, assembling the right team, attracting investors, and setting the machine in motion. With OneWeb, Wyler might stick around longer than he has at past ventures. Once he gets it working, assuming no catastrophes happen, he’ll complete the mission he gave himself after his mother died. “This is the second Internet,” he says. “It will be there for everybody.”