A tiny startup made a big splash this week by pledging to deliver a wireless Internet service fast enough to replace your home broadband connection.
But I wouldn't start drafting that nasty break-up letter to Comcast just yet.
Starry, a company based in New York and Boston, on Wednesday unveiled a 1-gigabit Internet service, which will launch with a trial in Boston over the summer. That's about 10 times faster than the average home Internet connection and speedy enough to let you download a two-hour high-definition movie in 25 seconds.
If the promise holds true, Starry could offer you a viable, potentially superior, alternative to the broadband service provided by your local phone or cable provider. The company believes it can undercut traditional broadband providers because it doesn't have to dig up streets to install physical lines. It would bring much needed competition and higher speeds. What's not to like?
It turns out that Starry's promise may be too good to be true. The service is based on a type of super-high radio frequency that can only travel over short distances and is easily distorted by walls or even rain. While this technology has been used to provide fat Internet pipes to businesses, its technical limitations make it expensive to deploy.
"It's extremely difficult to deliver a highly reliable, high-speed Internet service using these frequencies to residential customers," said Towerstream CEO Jeff Thompson. His company specializes in delivering wireless broadband to large buildings using microwave dishes and similar high-frequency radio waves.
But Starry CEO Chet Kanojia, best known for his failed attempt to shake up the broadcast industry by bringing live-streaming TV services to consumers through a company called Aereo, believes he's found the answer.
"The technologies we're using have been around for a long time and are dirt cheap," he said in an interview. "But no one else has stitched them together the way we have."
Starry is taking advantage of technological advances and off-the-shelf components used for Wi-Fi and satellite technology to work around these issues.
While a traditional provider installing cable or fiber for broadband might spend $2,500 for every home it hooks up, Starry can do it wirelessly for $25 per home, Kanojia said. That's thanks to a combination of tricks to amplify and direct the signal. He said the technologies allow him to transmit a signal 1 to 2 kilometers away.
Starry Beam radios will sit on top of buildings to capture the incoming signals. They're then transmitted to the Starry Point, another box that sits outside of the customer's window in their house or apartment, which then pipes the signal inside to yet another device called the Starry Wing. Starry Wing then broadcasts the connection throughout the home.
Starry hasn't offered pricing details on any of these products, but Kanojia said customers will not be responsible for the cost of the devices. Instead, pricing will be bundled into the service.
Dialing the right frequency
That all sounds great, but many are hung up on the airwaves that Starry plans to use.
Radio waves work like this: Higher frequency signals allow you to carry a lot of bandwidth, but at shorter ranges. That means you have to place more radios closer together. Wireless companies prefer to use lower band spectrum because they don't have to build as many cellular sites and because the signals can go through walls.
Starry is employing radio waves that range between 37 gigahertz and 39 gigahertz utilizing a technology called millimeter wavelength. At that high rate, walls and even heavy rainstorm or foggy mist can cause disruptions. In comparison, many of the US carriers use spectrum in the 700 megahertz band.
Some experts say that Starry is likely using the right technology to overcome technical hurdles, but they're skeptical the company can deliver the reliable performance at low costs.
Beyond technology, the bigger hurdle will be the cost to deploy the network, said Khurram Sheikh, president of SiBeam, which is working on chips that can tap into millimeter wavelength technology.
It's all about real estate
One of the biggest expenses will be finding enough rooftops for Starry to build a dense network. While the radios are relatively small and use little power, the company still needs to negotiate deals with building owners for access to rooftops and other places to put the equipment.
"All wireless providers are real estate plays," Thomson said. "The success of our business depends less on the technology and more on our ability to negotiate great long-term real estate deals."
Kanojia isn't too worried. He struck similar deals to get radios up for its Aereo service, which plucked broadcast TV signals and sent them down to its customers.
"That's just a process you have to go through," he quipped.
Many in the wireless community, especially those working on 5G, which also plans to use higher frequency radio signals, will be watching Starry closely.
"We like to see others innovating in this area," Sheikh said. "If a startup shows something real, it will accelerate things and we'll get more of this technology to market faster."
DISCUSS: STARRY'S CRAZY-FAST HOME INTERNET REPLACEMENT...
When Larry Page charges at you with a hockey stick, he doesn't talk trash. Known for being the silent type, Page is quiet and focused and lets his brother-in-arms, Sergey Brin, do the whooping.
That was the dynamic between the duo whenever they played roller hockey, a favorite pastime. But it's also what defines Google, the company they co-founded in 1998 as twentysomethings. Now the two have a bold new play. Their goal: to have their now-giant tech company feel once again like a scrappy startup, with Page focused on expanding the businesses and Brin touting their most ambitious projects.
Next week Page and Brin will give us our first real glimpse inside Alphabet, a new corporate structurethat makes several of the company's divisions, including the Google search engine itself, subsidiaries. Under the new organization, projects like Google's life sciences effort, called Verily, and smart-home device maker Nest become semi-independent companies. Alphabet reports financial earnings Monday. That's when the company will also share for the first time how some of its more experimental projects, so-called moon shots like a high-speed Internet service and delivery drones, have fared from a business perspective. The company, though, said Thursday that it won't be as detailed as some hoped.
For those of us who use Google's online apps and services, like Gmail and Google Docs, nothing really changes. The reorganization is mostly business jujitsu, intended to make investors happier by separating the money-making business engines from the projects still getting up to speed.
But their approach with Alphabet is actually a monumental move. Google is one of the most ambitious companies in the world. The tech it puts its brainpower behind can change industries, as was the case when Google announced it was working on self-driving cars in 2010. Automakers in Detroit initially scoffed. Now every major car company has an autonomous-driving initiative. That's what Google does to complacent industries: It charges at them, even if quietly, with a hockey stick.
"Sergey and I are seriously in the business of starting new things," Page said in a 976-word letter to the world when they unveiled the idea for Alphabet in August. "Our model is to have a strong CEO who runs each business, with Sergey and me in service to them as needed."
The decision to scramble Google, and to do it so gleefully as the company turns 18 this year, is classic Page and Brin. They've always done things a little differently, according to friends, mentors, business partners, employees and former employees interviewed for this story. Those Page and Brin fans say the lineage of this massive rewrite of Google can be traced to the early roller hockey days of their company.
Bruised and bloodied
To figure out what makes the iconic co-founders of Google tick, you need to go back to warm California nights at the turn of the millennium. In a nondescript parking lot outside Google's offices in Mountain View, cordoned off with police tape and traffic cones, Page, Brin and around 20 of their earliest employees played full-contact pickup roller hockey a few nights a week. They were decked out in official Google jerseys, with the company's colorful logo emblazoned on the front.
The matches weren't always gentle.
"People came back bruised and bloodied," remembers Doug Edwards, Google's former director of consumer marketing.
"Sergey was a little bit of a puck hog," recalls another early employee. That is, "unless Larry was playing, in which case, Sergey passed to him quite a lot. How's that for allegory?"
Now some 15 years later, the duo is charging forward again, with Page taking the puck. He'll lead as CEO of Alphabet, and his job will be to tap new leaders for its many businesses and make sure the company has its eyes focused on the future. Brin, Alphabet's president, serves as tech visionary, continuing in a role that's led him to push Google into self-driving cars, Wi-Fi-beaming balloons and smart eyewear.
Page and Brin, through spokespeople, declined requests to be interviewed.
But in his Alphabet letter, Page said, "From the start, we've always strived to do more, and to do important and meaningful things with the resources we have."
A careful read of his plans shows that the most important letters in Google's alphabet are DNA.
The user interface
Page, now 43, and Brin, 42, met in 1995 at an orientation at Stanford University, where they both studied computer science as grad students. But they were born worlds apart.
Page is a Michigan man. His grandfather was an assembly line worker for the Chevrolet plant in Flint, and his entire immediate family attended the University of Michigan. Brin, in contrast, was born in communist Moscow and came to the United States with his family when he was 6. His father was a professor at the University of Maryland, where Brin would go as an undergrad, and his mother worked as a research scientist at NASA's Goddard Space Flight Center.
Both Page and Brin credit their outrageous success to education, but not in the way you might think. They insist the key was Montessori schooling, which relies on a method that emphasizes collaborative learning without tests and grades. Other notable Montessori alumni are Amazon founder Jeff Bezos, Sean "P. Diddy" Combs and Golden State Warriors guard Stephen Curry, the NBA's reigning MVP.
"It was part of that training of not following rules and orders and being self-motivated," Page told TV's "20/20" in 2004. "Questioning what's going on in the world. Doing things a little bit differently."
Terry Winograd, Page's student adviser at Stanford, remembers the rule-breaking. Even after he started Google, Page would visit Winograd and glide into his office on skates. "All I remember is there was a sign at the front door that said 'No rollerblades in the halls,'" says Winograd.
That unique way of doing things meant Page was always a little enigmatic.
"He's quiet and nerdy. I mean that in a positive sense," says Edwards, who was employee No. 59 at the company and wrote a tell-all book called "I'm Feeling Lucky," about his time at Google from 1999 to 2005. "When I was in the room with him, my biggest goal was not to appear completely stupid."
It helped to have Eric Schmidt around. Schmidt, Google's CEO from 2001 to 2011 and now Alphabet's executive chairman, could read Page like no one else, Edwards says. "I refer to Eric as basically Larry's UI. He is a user interface by which you can understand what Larry is up to," Edwards says. "Because Larry is not very good at communicating that."
That doesn't mean Page doesn't like to have fun. At his wedding, Page and his wife, Lucy, went kite sailing into the sunset. And one year, for Brin's birthday, Page had Brin's office filled to the brim with giant exercise balls.
But if a Page prank is the exception, a Brin joke is the rule. Thanks to Brin, Google has a long tradition of pulling April Fools' pranks, like announcing in 2007 an Internet connection that's wired through your toilets, or 2005's Google Gulp, a drink that increases your intelligence. Brin once sent out emails telling employees he was leading a class for expectant mothers on the pillars of birthing, Edwards recalls. No one bought it.
Another time, Brin dressed up in a cow costume for Halloween and then went ahead and did an interview with a journalist. "I walked in to tell him the reporter was here, and Sergey was sitting there playing with his udders," recalls a former employee.
"Send him in," Brin replied.
On different planes
One day, more than a decade ago, Page emailed Sebastian Thrun out of the blue, asking if he wanted to see a robot he'd built. Thrun, a robotics expert from Germany, responded and joked about Page having the time to work on a project outside of Google's core mission. He was referring to Google's practice of allowing engineers to indulge curiosity, letting them devote 20 percent of work time to something other than a main project.
The two became friends. Then Page tried to tap Thrun for a project far wackier than anything Google had ever done: driverless cars. He asked Thrun, who had already done work on the subject, to take the lead.
Thrun said it couldn't be done.
"Please explain to me, in technical terms, why it can't be done," Page asked.
Thrun tried to put it into words, but couldn't. "I realized there's probably no explanation other than my lack of imagination," he recalls.
He took on the project, which became the genesis for Google X, the company's secretive research and development lab. Under Alphabet, the lab is its own company, simply known as X. In 2012, Thrun left his day-to-day role to start Udacity, an education startup, and then exited the company altogether in 2014.
Today, Alphabet has its hand in everything, from contact lenses equipped with computer chips that measure the glucose levels in your tears to balloons that fly around the stratosphere beaming Wi-Fi down to remote regions. People often contrast Google to its archrival, Apple, but when you dig deeper you see that they're not really the same. Apple is focused on its core consumer-device line of iPhones, iPads and Macs. Its late co-founder, Steve Jobs, once famously criticized Google for doing too much.
As for Jobs and Page, they're pretty different, though they both looked to legendary tech coach Bill Campbell as a mentor. (Campbell declined to comment.)
In fact, the two leaders are so opposite that "they are not even on the same planes of existence," says one person who worked in senior roles for both Apple and Google.
"There's enormous differences," he says. "Steve [was] a relentless driver, very in the moment. He [was] a great storyteller."
"Larry is not a storyteller. He's a futurist," he says. "His brain is always 10 or 20 years in the future."
It's that futurist mentality that attracts entrepreneurs to Google. And making sure those entrepreneurs can work in an environment that appeals to them is one of the main reasons for the Alphabet restructuring.
Google helped pioneer an acquisition strategy that's become the norm for tech companies. When the search giant bought YouTube in 2006, it vowed to let the budding video service keep some of its independence. Now YouTube is one of the brightest stars of Google's business.
The practice became a kind of Silicon Valley currency. Autonomy was as worthy a bargaining chip as stock options. Google bought Nest in 2013 for $3 billion, and it has become a model for how Alphabet companies should run.
Google also bought social-navigation app Waze in 2013 for nearly $1 billion. Co-founder Noam Bardin had Google's treatment of YouTube in mind when joining Google, he told CNET last year.
Waze has in some ways integrated itself into Google. While Nest has kept its offices separate, Waze moved its US headquarters from Palo Alto to Google's mothership in Mountain View. But Bardin insists Waze has stayed the same in every way that matters, especially when it comes to autonomy. "It was important for Larry to come and talk to us before the acquisition," Bardin said. "When he looks you in the eye and discusses it, it's from a fellow entrepreneur."
From the beginning
The Alphabet announcement wasn't the first time Page and Brin shocked the staid financial community.
When Google went public in 2004, the company did several things out of the ordinary. For starters, the initiative was called Project Denny's, because an early meeting over going public was so secret it was held at the roadside diner chain, said Lise Buyer, Google's former director of business optimization, who helped lead the company through the IPO.
The company also used a "Dutch" Internet auction to round up investors, instead of going the traditional route of having an investment bank do all the work buying and allocating shares. Page and Brin also wrote a 4,000-word manifesto on the kind of public company Google would be. From it, the company's famous mantra, "Don't be evil," was born.
Most notably, Google also used a stock structure that gave Page and Brin more control over the company than founders usually have.
A lot of those things, except the Dutch auction, have become pretty standard for other tech companies in the years since. But even Page and Brin couldn't push through all their ideas. One example: Google only wanted shareholders who knew what the business was doing, so the company wanted to require potential investors to take a five-question quiz before they would be allowed to buy. Nothing crazy -- simple questions like "What is the name of our advertising product?" to prove they were up to speed. The SEC didn't think a quiz like that was appropriate, Buyer recalls, so it didn't happen.
That's not the only thing the founders toned down. When, before the 2004 IPO, the two went to New York to pitch investors as part of what's commonly called a road show, people who attended some of those meetings said Page and Brin were arrogant and difficult when asked questions about the business. Plus, they were in casual Silicon Valley garb, out of place for buttoned-up Wall Street.
After the first meeting, the co-founders realized something was wrong and asked assistants to go out that night and buy them suits, says Buyer. "The next day, the turtlenecks were gone and they were wearing jackets."
For people who know Page and Brin, Alphabet wasn't a surprise. Most people interviewed for this story used the same word to describe it: inevitable.
"Larry had this vision from the very beginning," says Edwards. "It was never going to be just search. That's just where he started."
Page and Brin still don't wear suits if they don't need to. Page sports a puffy bubble jacket when it's cold, and Brin often wears athletic gear and Crocs. (Crocs didn't respond to multiple requests for comment regarding Brin's affinity for the footwear.)
They've come a long way from those Google hockey jerseys. But also, not really.
Says Winograd, Page's adviser at Stanford: "They always had the sense of, 'Let's try something, and if it sounds crazy, maybe it's worth trying. We'll find out.'"
DISCUSS: ALPHABET? GOOGLE? EITHER WAY, IT'S READY...