In the Plex: How Google Thinks, Works, and Shapes Our Lives (Steven Levy, 2011)

I read this book recently. Going into this, I knew I wouldn't learn many new/surprising things. After all, I have been following Google from outside since 1999. After I saw Google in action, I had switched to it immediately. I was quick to get on board with Gmail as well, and followed many Google services over the years. I have seen things. 

As I anticipated, I didn't find many surprises in the book. But I liked the book nevertheless. It is always nice to read stuff from Steven Levy. 

Steven Levy is one of my favorite authors. I have a dog-eared copy of the Hackers book he wrote. It was excellent writing. So much research went into each character in the book, and he made each character come alive. He also weaved the narration together artistically around a handful of themes and related new information/events back to these themes. He had put a lot of his interpretation and insight in to the facts/narrative, and I really enjoyed that. 

This book is not as colorful. It has a raw bland reporting kind of narration. If the author name was obscured, I would guess it came from a competent author, but not necessarily from Steven Levy. The Facebook book by Levy is better. (I am halfway through it now.) That one feels more personal and intimate.  It has Levy's color and style over it.

One thing I wished Levy did in this book would be to explore more and comment about what led to so many botched up products at Google. What made Google incompetent about productizing things? How much of it has to do with bad management due to big size, and how much of it is due to a culture problem? 

Through reading the book I got a new appreciation for the importance of managing for the success of a company, and how hard it is to manage teams and products. Good intentions don't work, good systems work, and it is hard to implement good systems for scaling innovation and product creation. The Creativity Inc book had some good discussion about managing the creative process, and I wish this book could also explore more in to the dynamics at play at Google in terms of project and product management. 

I think the book was overly nice to Google and avoided any serious criticism. 

Google used some shoddy tactics in negotiating tax reliefs and for imposing aggressive terms when opening datacenters. (Part 4, Google's Cloud chapter was my favorite chapter in the book.) Moreover Google was also involved in several privacy violations as well. Don't be evil is a low bar to clear in the first place, but they had to remove that in 2018 update to Google Code of Conduct. 

Ok, here are my highlights from the book. It is long, but worth scanning over. Remember again this book is from 2011, so it doesn't cover the developments after that point.


The science fiction writer William Gibson once said that the future is already here --just not evenly distributed. At Google, the future is already under way. To understand this pioneering company and its people is to grasp our technological destiny. And so here is Google: how it works, what it thinks, why it’s changing, how it will continue to change us. And how it hopes to maintain its soul.

By all metrics, the company was still thriving. Google still retained its hundreds of millions of users, hosted billions of searches every day, and had growing businesses in video and wireless devices. Its employees were still idealistic and ambitious in the best sense. But a shadow now darkened Google’s image. To many outsiders, the corporate motto that Google had taken seriously — “ Don’t be evil ” — had become a joke, a bludgeon to be used against it. What had happened ?

Part One: The World According to Google: Biography of a Search Engine

Page was not a social animal — people who talked to him often wondered if there were a jigger of Asperger’s in the mix — and could unnerve people by simply not talking. But when he did speak, more often than not he would come out with ideas that bordered on the fantastic.

He believed that the only true failure was not attempting the audacious. "Even if you fail at your ambitious thing, it’s very hard to fail completely." 

When people proposed a short - term solution, Page’s instinct was to think long term. 

Page chose to work in the [Stanford] department’s Human - Computer Interaction Group. The subject would stand Page in good stead in the future with respect to product development, even though it was not in the HCI domain to figure out a new model of information retrieval. On his desk and permeating his conversations was Apple interface guru Donald Norman’s classic tome The Psychology of Everyday Things.

"I thought he was pretty obnoxious," Page later said of the guide, Sergey Brin. The content of the encounter is now relegated to legend, but their argumentative banter was almost certainly good - natured. Despite the contrast in personalities, in some ways they were twins. Both felt most comfortable in the meritocracy of academia, where brains trumped everything else. Both had an innate understanding of how the ultraconnected world that they enjoyed as computer science (CS) students was about to spread throughout society. Both shared a core belief in the primacy of data. And both were rock stubborn when it came to pursuing their beliefs. When Page settled in that September, he became close friends with Brin, to the point where people thought of them as a set : LarryAndSergey.

He had arrived at Stanford at nineteen after whizzing through the University of Maryland, where his father taught, in three years; he was one of the youngest students ever to start the Stanford PhD program. 

Donald Knuth, a Stanford professor whose magisterial series of books on the art of computer programming made him the Proust of computer code, recalls driving down the Pacific coast to a conference with Sergey one afternoon and being impressed at his grasp of complicated issues. His adviser, Hector Garcia - Molina, had seen a lot of bright kids go through Stanford, but Brin stood out. “He was brilliant,” Garcia-Molina says.

Having a human being determine the ratings was out of the question. First, it was inherently impractical. Further, humans were unreliable. Only algorithms — well drawn, efficiently executed, and based on sound data — could deliver unbiased results. So the problem became finding the right data to determine whose comments were more trustworthy, or interesting, than others. Page realized that such data already existed and no one else was really using it. He asked Brin, “ Why don’t we use the links on the web to do that ? ” Page, a child of academia, understood that web links were like citations in a scholarly article. It was widely recognized that you could identify which papers were really important without reading them — simply tally up how many other papers cited them in notes and bibliographies.

Every other week Page would come to Garcia - Molina’s office asking for disks and equipment. “ That’s fine, ” Garcia - Molina would say. “ This is a great project, but you need to give me a budget. ” He asked Page to pick a number, to say how much of the web he needed to crawl, and to estimate how many disks that would take. “ I want to crawl the whole web, ” Page said. Page indulged in a little vanity in naming the part of the system that rated websites by the incoming links : he called it PageRank. But it was a sly vanity ; many people assumed the name referred to web pages, not a surname.

Hassan was also good friends with Brin, whom he’d met at an Ultimate Frisbee game during his first week at Stanford. Page’s program “ had so many bugs in it, it wasn’t funny, ” says Hassan. Part of the problem was that Page was using the relatively new computer language Java for his ambitious project, and Java kept crashing. “ I went and tried to fix some of the bugs in Java itself, and after doing this ten times, I decided it was a waste of time, ” says Hassan.

He wrote a program in Python — a more flexible language that was becoming popular for web - based programs — that would act as a “ spider, ” so called because it would crawl the web for data. The program would visit a web page, find all the links, and put them into a queue. Then it would check to see if it had visited those link pages previously. If it hadn’t, it would put the link on a queue of future destinations to visit and repeat the process. Since Page wasn’t familiar with Python, Hassan became a member of the team. He and another student, Alan Steremberg, became paid assistants to the project. Brin, the math prodigy, took on the huge task of crunching the mathematics that would make sense of the mess of links uncovered by their monster survey of the growing web.

“Even the first set of results was very convincing,” Hector Garcia - Molina says. “It was pretty clear to everyone who saw this demo that this was a very good, very powerful way to order things.” “We realized it worked really, really well,” says Page. “And I said, ‘Wow, the big problem here is not annotation. We should now use it not just for ranking annotations, but for ranking searches.’ ” It seemed the obvious application for an invention that gave a ranking to every page on the web. “It was pretty clear to me and the rest of the group,” he says, “that if you have a way of ranking things based not just on the page itself but based on what the world thought of that page, that would be a really valuable thing for search.”

AltaVista analyzed what was on each individual page — using metrics like how many times each word appeared — to see if a page was a relevant match to a given keyword in a query.

“The whole field had suffered blinders,” says the computer scientist Amit Singhal, then a Bell Labs researcher who had been a protégé of Jerry Salton. “In some sense, search really did need two people who were never tainted by people like me to come up with that shake - up.”

Kleinberg began showing his breakthrough around IBM. His managers quickly put him in touch with the patent lawyers. Most people took a look at what Kleinberg had set up and wanted him to find stuff for them. Even the patent attorney wanted Kleinberg to help him find sources for his hobby, medieval siege devices. By February 1997, he says, “ all sorts of IBM vice presidents were trooping through Almaden to look at demos of this thing and trying to think about what they could do with it. ” Ultimately, the answer was … not much. IBM was a $ 70 billion business, and it was hard to see how a research project about links on this World Wide Web could make a difference. Kleinberg shrugged it off. He was going to teach computer science at Cornell.

There was yet a third person with the idea, a Chinese engineer named Yanhong ( Robin ) Li. In 1987, he began his studies at Beijing University, an institution that claimed prominence in the country by way of a metric : The Science Citation Index, which ranked scientific papers by the number of other papers that cited them.

Li came to the United States in 1991 to get a master’s degree at SUNY Buffalo, and in 1994 took a job at IDD Information Services in Scotch Plains, New Jersey, a division of Dow Jones. Part of his job was improving information retrieval processes. He tried the search engines at the time — AltaVista, Excite, Lycos — and found them ineffectual and spam - ridden. One day in April 1996 he was at an academic conference. Bored by the presentation, he began to ponder how search engines could be improved. He realized that the Science Citation Index phenomenon could be applied to the Internet. The hypertext link could be regarded as a citation! “ When I returned home, I started to write this down and realized it was revolutionary, ” he says. He devised a search approach that calculated relevance from both the frequency of links and the content of anchor text. He called his system RankDex.

When he described his scheme to his boss at Dow Jones, urging the company to apply for a patent, he was at first encouraged, then disappointed when nothing happened. “ So a couple of months later, I decided to write the application by myself. ” He bought a self - help book on patent applications and filed his in June 1996. But when he told his boss, Dow Jones reasserted itself and hired a lawyer to review the patent, which it refiled in February 1997. ( Stanford University would not file its patent for Larry Page’s PageRank system until January 1998. ) Nonetheless, Dow Jones did nothing with Li’s system. “ I tried to convince them it was important, but their business had nothing to do with Internet search, so they didn’t care, ” he says.

Robin Li quit and joined the West Coast search company called Info - seek. In 1999, Disney bought the company and soon thereafter Li returned to China. It was there in Beijing that he would later meet — and compete with — Larry Page and Sergey Brin.

Cheriton himself was a prime example of how the Stanford network launched companies and enriched the founders. 

Though Brin and Page didn’t have classes with Cheriton, they headed to his office for some advice. They specifically wanted to know how they might interest a company into using PageRank in its own search technology. Cheriton told them that it would be difficult — Sun Microsystems, he reminded them, had been started out of frustration when companies had spurned Bechtolsheim’s attempts to sell his workstation technology. Yet Brin and Page were reluctant at that point to strike out on their own. They had both headed to Stanford intending to become PhDs like their dads.

Extrapolating that in terms of increased ad revenue, Excite would take in $13,000 more every day, for a total of $47 million in a year. Page envisioned his tenure at Excite lasting for seven months, long enough to help the company implement the search engine. Then he would leave, in time for the fall 1997 Stanford semester, resuming his progress toward a doctorate. Excite’s total outlay would be $1.6 million, including $30,000 to Stanford for the license, a $20,000 salary, a $40,000 bonus for implementing it within three months, and $ 70,000 in Excite stock.

Bell was visibly upset. The Stanford product was too good. If Excite were to host a search engine that instantly gave people information they sought, he explained, the users would leave the site instantly. Since his ad revenue came from people staying on the site — “ stickiness ” was the most desired metric in websites at the time — using BackRub’s technology would be counterproductive. “ He told us he wanted Excite’s search engine to be 80 percent as good as the other search engines, ” says Hassan. And we were like, “ Wow, these guys don’t know what they’re talking about. ”

Hassan says that he urged Larry and Sergey right then, in early 1997, to leave Stanford and start a company. “Everybody else was doing it,” he says. “I saw Hotmail and Netscape doing really well. Money was flowing into the Valley. So I said to them, ‘ The search engine is the idea. We should do this. ’ They didn’t think so. Larry and Sergey were both very adamant that they could build this search engine at Stanford.”

But for the next year and a half, all the companies they approached turned them down. “ We couldn’t get anyone interested, ” says Page. “ We did get offers, but they weren’t for much money.

They went back to Dave Cheriton, who encouraged them to just get going. “ Money shouldn’t be a problem, ” he said. Cheriton suggested that they meet with Andy Bechtolsheim. Brin dashed off an email to Bechtolsheim that evening around midnight and got an immediate reply asking if the two students could show up at eight the next morning at Cheriton’s house, which was on the route Bechtolsheim used to go to work each day.

Bechtolsheim, impressed but eager to get to the office, cut the meeting short by offering to write the duo a $10,000 check. “We don’t have a bank account yet,” said Brin. “Deposit it when you get one,” said Bechtolsheim, who raced off in his Porsche. With as little fanfare as if he were grabbing a latte on the way to work, he had just invested in an enterprise that would change the way the world accessed information. Brin and Page celebrated with a Burger King breakfast. The check remained in Page’s dorm room for a month.

After the Bechtolsheim meeting, Shriram invited them to his house to meet his boss Jeff Bezos, who was enthralled with their passion and “ healthy stubbornness, ” as they explained why they would never put display ads on their home page. Bezos joined Bechtolsheim, Cheriton, and Shriram as investors, making for a total of a million dollars of angel money.

On September 4, 1998, Page and Brin filed for incorporation and finally moved off campus. Sergey’s girlfriend at the time was friendly with a manager at Intel named Susan Wojcicki, who had just purchased a house on Santa Margarita Street in Menlo Park with her husband for $ 61,000. To help meet the mortgage, the couple charged Google $ ,700 a month to rent the garage and several rooms in the house. At that point they’d taken on their first employee, fellow Stanford student Craig Silverstein. He’d originally connected with them by offering to show them a way to compress all the crawled links so they could be stored in memory and run faster. ( “ It was basically to get my foot in the door, ” he says. ) They also hired an office manager. But almost as if they were still hedging on their PhDs, they maintained a presence at Stanford that fall, coteaching a course, CS 349, “ Data Mining, Search, and the World Wide Web, ” which met twice a week that semester. Brin and Page announced it as a “ project class ” in which the students would work with the repository of 25 million web pages that they had captured as part of what was now a private company. They even had a research assistant. The first assigned reading was their own paper, but later in the semester a class was devoted to a comparison of PageRank and Kleinberg’s work.

“ When we started the company, we had two computers, ” says Craig Silverstein. “ One was the web service, and one was doing everything else — the page rank, the searches. And there was a giant chain of disks that went off the back of the computer that stored twenty - five million web pages. Obviously that was not going to scale very well. ” Getting more computers was no problem. Google needed brainpower, especially since Brin and Page had reached the limits of what they could do in writing the software that would enable the search engine to grow and improve.

Some of Google’s early hires were simply brainy recent grads, people like Marissa Mayer, a hard - driving math whiz and ballet dancer in her high school in Wausau, Wisconsin, who had become an artificial intelligence star at Stanford. ( During her interview with Silverstein, she was asked for three things Google could do better ; ten years later, she was still kicking herself that she listed only two. )

One of their first coups was a professor at the University of California at Santa Barbara named Urs Hölzle.

Part of the problem was that Page and Brin had written the system in what Hölzle calls “ university code, ” a nice way of saying amateurish. “ The web server couldn’t handle more than ten requests or so a second because it was written in Python, which is a great idea for a research system, but it’s not a high - performance solution, ” he says. He immediately set about rewriting the code.

One of the DEC engineers had already independently discovered the power of web links in search. Jeffrey Dean suspected that it would be helpful to web users if a software program could point them to pages that were related to the ones they liked. In his vision, you would be reading an article in The New York Times and his program would pop up, asking if you’d like to see ten other interesting pages related to the one you were reading.

In February 1999, Dean bailed from DEC to join a start - up called mySimon. Within a few months, though, he was bored. Then he heard that Urs Hölzle, whom he’d known through his grad school adviser, had joined up with the guys who did PageRank. “ I figured Google would be better because I knew more of the people there, and they seemed like they were more technically savvy, ” he says. He was so excited about working there that even though his official starting date wasn’t until August 1999, in July he began coming to Google after his workday at mySimon ended.

The experience led to an ambitious revamp of the way the entire Google infrastructure dealt with files. “ I always had wanted to build a file system, and it was pretty clear that this was something we were going to have to do, ” says [Sanjay] Ghemawat, who led the team.

The response of Yahoo’s users to the Google technology, though, was probably more conscious. They noticed that search was better and used it more. “ It increased traffic by, like, 50 percent in two months, ” Manber recalls of the switch to Google. But the only comment he got from Yahoo executives was complaints that people were searching too much and they would have to pay higher fees to Google.

Sergey Brin had written the original part of the Google search engine that dealt with relevance. At that point it was largely based on PageRank, but as early as 2000 Amit Singhal realized that as time went on, more and more interpretive signals would be added, making PageRank a diminishing factor in determining results. ( Indeed, by 2009, Google would say it made use of more than two hundred signals — though the real number was almost certainly much more — including synonyms, geographic signals, freshness signals, and even a signal for websites selling pizzas. )

Singhal completed a version of the new code in two months and by January 2001 was testing it. Over the next few months, Google exposed it to a percentage of its users and liked the results. They were happier. Sometime that summer, Google flipped the switch and became a different, more accurate service. In accordance with the company’s fanatical secrecy on such matters, it made no announcement. Five years later, Singhal was acknowledged by being named a Google Fellow, awarded an undisclosed prize that was almost certainly in the millions of dollars.

All this helped to explain how Google could find someone whose name may have never appeared in a search before. (One - third of all search queries are virgin requests.)

Consider that these were geeky mathematicians who in an earlier era would have written obscure papers and be scraping by financially on an academic’s salary. Now their work directly benefited hundreds of millions of people, and they had in some way changed the world. Plus, many of them owned stock options that had made them very wealthy.

Part Two: Googlenomics: Cracking the Code on Internet Profits

Brin and Page wanted to work with Moritz. But they also wanted to work with Doerr.

It was not a choice that either Doerr or Moritz would have preferred. But both VCs recognized Google as perhaps the last big score of the Internet boom, so they agreed to the unusual arrangement, splitting the $ 25 million of capital that the company required.

Larry began by explaining Google’s recently refined mission : “ To organize the world’s information, making it universally accessible and useful. ” He talked about Google using artificial intelligence and having a million computers someday. None of this was surprising to the reporters. Start - up founders talked like that all the time. How could the press know that this was the one time when the fantastic predictions would be realized ?

From Kleiner Perkins came a recommendation for a thirty - five - year - old Iran - born executive named Omid Kordestani.

Kordestani, never at a loss in social situations, and he began to talk about what qualities they might consider in a vice president of business operations. Brin called in Urs Hölzle and everyone else still hanging around the office. They all went out for dinner at the Mandarin Gourmet in Palo Alto. Kordestani picked up the tab — not a bad investment, considering that the stake he was granted by accepting the job at Google would be worth $ 2 billion within a decade.

“ Marketing was always the poor stepchild at Google, because Larry and Sergey really thought you can build a company without it, ”

The fact was, the Google search engine marketed itself. As people discovered novel ways to use it, the company name became a verb.

On the first day of the deal, early arrivals at headquarters discovered that there weren’t enough servers to run searches on both Google and the Netscape home page. So Google turned off its own home page — stranding its loyal home page users — until it could get more servers.

Google’s first stab at selling advertising began in July 1999. When Jeff Dean arrived from DEC — a couple of months before he toiled in the war room to fix the indexing problem — Brin and Page told him that they needed an ad system. But they had no idea what a Google ad should be.

Page and Brin agreed to Doerr’s Magical Mystery Tour of high - tech royalty : Apple’s Steve Jobs, Intel’s Andy Grove, Intuit’s Scott Cook, Amazon’s Jeff Bezos, and others. Then they came back to Doerr. “ This may surprise you, ” they told him, “ but we agree with you. ” They were ready to hire a CEO.

Schmidt, then forty - six, had been the chief technology officer at Sun Microsystems and was the CEO of the big networking company Novell. He was familiar with boardrooms and bottom lines. But the big factor in his favor was that he was an excellent engineer, with a Berkeley computer science PhD and geek renown as the coauthor of lex, a coding tool that was beloved by hard - core UNIX programmers.

Schmidt adopted a public stance toward the founders of unfettered admiration, a position he carefully maintained thereafter. “ I fairly quickly figured out these guys are good at what they do, ” he told me in early 2002. “ Sergey is the soul and the conscience of the business. He’s a showman who cares deeply about the culture, the one who talks more, with a bit of Johnny Carson. Larry is the brilliant inventor, the Edison. Every day I am thankful I accepted this job offer. ”

“ He kind of came in here like a visiting professor, not the classic CEO with command and control, ” says Omid Kordestani.

As late as 2002, the founders still sounded bitter when explaining why Schmidt was hired. “ Basically, we needed adult supervision, ” said Brin, adding that their VC investors “ feel more comfortable with us now — what do they think two hooligans are going to do with their millions ? ” The transition was rocky, but as the years went by, Page and Brin seemed to genuinely appreciate Schmidt’s contribution. Page would come to describe the CEO’s hiring as “ brilliant. ”

What did Patel learn about being CEO ? “ Anything that’s wrong sort of bubbles up, so you have to deal with all these problems that aren’t the sort of problems I would want to solve, ” he says. “ It’s not a job I would want. ” He had a better job, anyway. He was an engineer at Google.

Then came a development that was sudden, transforming, decisive, and, for Google’s investors and employees, glorious. Google launched the most successful scheme for making money on the Internet that the world had ever seen. More than a decade after its launch, it is nowhere near being matched by any competitor. It became the lifeblood of Google, funding every new idea and innovation the company conceived of thereafter. It was called AdWords, and soon after its appearance, Google’s money problems were over. Google began making so much money that its biggest problem was hiding how much.

One of the biggest was the adoption of the other Overture idea, pay per click.

The longtime joke in the marketing world was that only half of advertising is worth the money — but no one can tell which half. Google was switching the game : using its system, you would pay for ads only when they worked.

Here was the rub : The bids submitted by contenders for the ad slots were only half of what ultimately determined the winners of the auction. The other half was the quality score. This metric would assure that the ads Google showed on its results page were helpful to its users — a high quality score meant that the ad was relevant to the user’s quest. Low quality scores were for ads that were irrelevant, misleading, or even spamlike.

Later Google used a more complicated formula to determine quality score by adding factors such as the relevance of the ad to the specific keyword and the quality of the landing page. But the biggest factor remained the predicted click - through rate.

The system enforced Google’s insistence that advertising shouldn’t be a transaction between publisher and advertiser but a three - way relationship that also included the user.

The auction would be conducted in stealth, generated the instant someone typed a keyword into the Google search box, with the result shown in a fraction of a second.

"I don’t know the number of auctions that we run per day, but for purposes of argument, use a billion or a hundred million, ” says Schmidt.

From that point on, revenue from the right - hand side of Google’s search results page — which had previously constituted only 10 to 15 percent of Google’s ad take, with the bulk coming from the direct sales of premium ads — began rising.

Every time the test was run, the outcome was similar : dropping ads did not increase searching. More often than not, the users in the control group who continued to see ads searched more than those with ad - free pages. Google’s relieved conclusion : its ads made people happy. )

Google innovated circles around Overture, focusing on its core obsessions of speed and scale. Overture required its advertisers to pick specific keywords ; Google would match an ad to many keywords, some of them with subtle connections discovered by analysis of the behavior of its millions of users.

Google built a self - service system that allowed it to accommodate hundreds of thousands of advertisers.

Part Three: Don’t Be Evil: How Google Built Its Culture

You can’t understand Google, ” she said, “ unless you know that both Larry and Sergey were Montessori kids. ”

It’s almost as if Larry and Sergey were thinking of Maria Montessori’s claim “ Discipline must come through liberty …. We do not consider an individual disciplined only when he has been rendered as artificially silent as a mute and as immovable as a paralytic. He is an individual annihilated, not disciplined. We call an individual disciplined when he is master of himself. ” Just as it was crucial to Montessori that nothing a teacher does destroy a child’s creative innocence, Brin and Page felt that Google’s leaders should not annihilate an engineer’s impulse to change the world by coding up some kind of moon shot. “ We designed Google, ” Urs Hölzle says, “ to be the kind of place where the kind of people we wanted to work here would work for free. ”

The shares they purchased for $ 85 realized an 18 percent profit in a single day. Though Wall Street had gotten its licks in, Google could claim success insofar as the auction process gave equal access to all investors. None of that mattered in the ensuing months and years as the stock took off. The stock price climbed to $ 280 a year later, $ 383 a year after that, and a little more than three years after the IPO, topped $700.

Page and Brin wanted it all : a company with thousands of engineers that ran smoothly while still indulging the creative impulses of its people. Every time the head count doubled, the question came up again : could Google’s bottom - up style of management actually scale ? Page and Brin never doubted it.

Brin and Page came up with a solution : Google would no longer have managers. At least not in engineering. Instead, they figured, the engineers could self - organize. That approach worked well in the nascent days of Google. 

Stacy Sullivan, the head of HR, begged Page and Brin not to go through with it. “ You can’t just self - organize ! ” she told them. “ People need someone to go to when they have problems ! ” The newly arrived Schmidt and the company’s unofficial executive coach, Bill Campbell, weren’t happy with the idea, either.

One by one, Campbell and Page summoned them in, and one by one Page asked them, “ Do you want to be managed ? ” As Campbell would later recall, “ Everyone said yeah. ” Page wanted to know why. They told him they wanted somebody to learn from. When they disagreed with colleagues and discussions reached an impasse, they needed someone who could break the ties.

Google was taking on new engineers at a furious rate, and, brilliant as they were, the new people needed some guidance to figure out what to do.

On Rakowski’s first day, she met him in the Ping - Pong conference room. “ You’re going to be working on Gmail, ” she said. Rakowski was speechless. “ I was twenty - two years old, ” he says. “ I was shocked that they were going to let someone that young and inexperienced do that job. ”

A product manager at Google didn’t give orders. His ( or her ) job was to charm the engineers into a certain way of thinking. It was a Mensa form of cat herding.

As the years passed and Google’s management system became formalized, a corporate amnesia seemed to envelop Brin and Page’s 2001 kill - the - managers caper.

Doerr was a fan of a complicated system called Objectives and Key Results, usually referred to by the acronym OKR. It was something Andy Grove had devised at Intel ( he’d called it Management by Objective ), but Doerr believed it was even more useful for start - ups.

OKRs became an essential component of Google culture. Every employee had to set, and then get approval for, quarterly OKRs and annual OKRs. There were OKRs at the team level, the department level, and even the company level.

Four times a year, everything stopped at Google for divisionwide meetings to assess OKR progress.

An employee didn’t say, “ I will make Gmail a success ” but, “ I will launch Gmail in September and have a million users by November. ” “ It’s not a key result unless it has a number, ” says Marissa Mayer.

Googlers looking for a cool new project could access a section called simply “ Ideas, ” where their colleagues pitched promising concepts that needed manpower.

The company was an information lobster, hard-shelled on the outside but soft and accessible on the inside.

Whatever they felt was important at the moment would be their work. Sergey sometimes liked to move his workplace right in the middle of a project he found interesting.

“ Larry can be a very, very sensitive and good person, ” says one former Google PR hand, “ but he has major trust issues and few social graces. Sergey has social graces, but he doesn’t trust people who he thinks don’t approach his level of intelligence. ”

Part Four: Google’s Cloud: Building Data Centers That Hold Everything Ever Written

So he unilaterally decided to revisit his college project and create a web - based email product for Google. He would have the benefit of something that was only just invented when he was at Case Western : the dynamic JavaScript computing language, which could, if you pushed it, help you create a web - based application that behaved like a desktop application running directly on the computer.

But what would really distinguish Buchheit’s email product from its competitors was storage.

His mailboxes would have a gigabyte of storage — more than a hundred times what competitors offered. For free.

“ Almost everyone thought it was a bad idea, ” says Buchheit. “ Except Larry and Sergey. ”

Larry and Sergey liked Caribou too much to kill it but dished out very tough love. At one point Page told the group, “ I’d rather be doused with gasoline and set on fire than use your product. ” But finally it was ready to be released in beta version.

But it wasn’t just Gmail that disturbed people. Suddenly, Google itself was suspect. Until 2004 Google had been seen as a feisty start - up performing an invaluable service. But it was sitting on a privacy tinderbox. One key issue was Google’s retention of user requests and responses when they visited its search engine. You can’t get more personal than that. A search history can reveal your health problems, your commercial interests, your hobbies, and your dreams. What would your health care insurer think about your search for “ chest pains ” ? What would your investors think if you searched for “ bankruptcy lawyers ” ? What would the cops think if you searched for “ hydroponic equipment ” ? What would your spouse think if you searched for “ afternoon sex encounters ” ? What would the government think if you searched for “ tax resistance ” ? In 2006, the government, in a fishing expedition for information to help efforts to regulate pornography, would demand that Google and other search engines hand over logs of millions of searches. Google alone fought the subpoena. But when privacy advocates demanded that Google not retain any logs at all, the company balked. Those logs were the lifeblood of Google’s persistent drive to improve itself, the oxygen of its effort to become an unprecedented learning machine.

Then in July 2005, a CNET reporter used Schmidt as an example of how much personal information Google search could expose. Though she used only information that anyone would see if they typed Schmidt’s name into his company’s search box, Schmidt was so furious that he blackballed the news organization for a year. “ My personal view is that private information that is really private, you should be able to delete from history, ” Schmidt once said. But that wasn’t Google’s policy. If Google’s own CEO had trouble dealing with privacy, how could ordinary people cope?

Gates looked stunned, as if this offended him. “ How could you need more than a gig ? ” he asked. “ What’ve you got in there ? Movies ? PowerPoint presentations ? ” No, just lots of mail. He began firing questions. “ How many messages are there ? ” he demanded. “ Seriously, I’m trying to understand whether it’s the number of messages or the size of messages. ” After doing the math in his head, he came to the conclusion that Google was doing something wrong.

The episode is telling. Gates’s implicit criticism of Gmail was that it was wasteful in its means of storing each email. Despite his currency with cutting - edge technologies, his mentality was anchored in the old paradigm of storage being a commodity that must be conserved. He had written his first programs under a brutal imperative for brevity. And Microsoft’s web - based email service reflected that parsimony.\

The term originally wasn’t popular at Google. “ Internally, we thought of ‘ cloud computing ’ as a marketing term, ” says Urs Hölzle. ( “ Marketing ” being pejorative in this context. ) “ Technically speaking, it’s cluster computing that you do. ”

In 2002, before Google firmly closed the shutters, I was offered a rare glimpse of the company’s data storage. Jim Reese, then the caretaker of the company’s infrastructure, was the guide. He drove to the Exodus colo ( colocation center ) near San Jose in his car, apologizing for a flapping patch of upholstery on the interior roof as he steered. On the way over, he shared the kind of information that in later years Google would never divulge : real numbers about its servers and its searches. Google, he said, had 1,000 servers to process the 150 million searches its customers launched every day. A sleepy guard waved us in, and we entered a large darkened space with “ cages ” of servers surrounded by chain - link fences. Air conditioners churned out a steady electronic hum. Reese pointed out who owned the servers in each cage.

Google was a tough client for Exodus ; no company had ever jammed so many servers into so small an area. The typical practice was to put between five and ten servers on a rack ; Google managed to get eighty servers on each of its racks. The racks were so closely arranged that it was difficult for a human being to squeeze into the aisle between them. To get an extra rack in, Google had to get Exodus to temporarily remove the side wall of the cage. “ The data centers had never worried about how much power and AC went into each cage, because it was never close to being maxed out, ” says Reese. “ Well, we completely maxed out. It was on an order of magnitude of a small suburban neighborhood, ” Reese says. Exodus had to scramble to install heavier circuitry. Its air - conditioning was also overwhelmed, and the colo bought a portable AC truck. They drove the eighteen - wheeler up to the colo, punched three holes in the wall, and pumped cold air into Google’s cage through PVC pipes.

“ They told me that whatever I do, make sure it will work not just for 500 or ,000 computers but 5,000 — that we should build in massive scalability now and that we would have that many computers in just a few years. Which we did, ” says Reese.

The key to Google’s efficiency was buying low - quality equipment dirt cheap and applying brainpower to work around the inevitably high failure rate.

Buchheit remembers one time when he was doing an early Gmail demo in Larry’s office. Page made a face and told him it was way too slow. Buchheit objected, but Page reiterated his complaint, charging that the reload took at least 600 milliseconds. ( That’s six - tenths of a second. ) Buchheit thought, You can’t know that, but when he got back to his own office he checked the server logs. Six hundred milliseconds. “ He nailed it, ” says Buchheit. “ So I started testing myself, and without too much effort, I could estimate times to a hundred milliseconds precision — I could tell if it was 300 milliseconds or 700, whatever. And that happens throughout the company. ” 

In 2008, Google issued a Code Yellow for speed. ( A Code Yellow is named after a tank top of that color owned by engineering director Wayne Rosing. During Code Yellow a leader is given the shirt and can tap anyone at Google and force him or her to drop a current project to help out. Often, the Code Yellow leader escalates the emergency into a war room situation and pulls people out of their offices and into a conference room for a more extended struggle. )

To help meet its goals, the company created a market - based incentive program for product teams to juice up performance — a cap - and - trade model in which teams were mandated latency ceilings or maximum performance times. If a team didn’t make its benchmarks, says Hölzle, it accrued a debt that had to be paid off by barter with a team that exceeded its benchmarks. “ You could trade for an engineer or machines. Whatever, ” he says.

The tricky part of setting up in a new facility was loading all those thousands of servers with the indexes. That involved terabytes of data, which was potentially going to force Google to pay a huge amount of money to the bandwidth provider that owned the fiber. “ Networking was very expensive, ” says Hölzle. “ And our data push would take twenty hours at a gigabyte per second — that would cost us something like $ 25,000 a month. ” To save money, Google devised a trick that exploited a loophole in the billing system for data transfer. Broadband providers used a system known as the 95th Percentile Rule. Over the period of a month, the provider would test how much information was moving, automatically taking a measurement every five minutes. In order to discard unusual spikes in activity, when the billing rate was calculated the provider would lop off the measurements in the top five percentiles and bill the customer at the rate of the 95th percentile. Google’s exploitation of the rule was like the correct answer to a trick question in one of its hiring interviews. It decided to move all its information during those discounted spikes. “ We figured out that if we used zero bandwidth all month, except for thirty hours once a month, we would be under that 5 percent, ” says Reese. For two nights a month, from 6 p.m. to 6 a.m. Pacific time, Google moved all the data in its indexes from west to east.

And of course, the bill came out to be nothing, ” says Reese, “ because when they lopped off the top 5 percent, our remaining bandwidth was in fact zero, because we didn’t use any otherwise. I literally turned off the router ports for twenty - eight or twenty - nine days a month. ”

Fiber - optic cable was the most efficient, robust, and speedy means of moving data. Just as Google had taken advantage of the oversupply of data centers in the wake of the dot - com bust, it had a great opportunity to buy fiber - optic cable cheap. In the 1980s and 1990s, a raft of optical networking companies had made huge investments in fiber optics. But they had overestimated the demand, and by the early 2000s, many were struggling or going broke. Google began buying strategically located stretches of fiber. “ We would want to pick up pieces that would connect our data center, so we’d identify the owner, negotiate, and take it over, ” says Chris Sacca, who did many of the deals. “ Then we’d put optical networking equipment on one end in our data center, the same equipment on the data center at the other end, and now we’re running that stretch of fiber, ” says Sacca. “ We were paying ten cents on the dollar. ” Since fiber - optic cable had huge capacity, Google then made arrangements with broadband companies to fill in the gaps it didn’t own. “ We swapped out strands with other guys, ” says Sacca. By the time Google finished with its fiber push, it was in a unique situation. “ We owned the fiber. It was ours. Pushing the traffic was nothing, ” says Sacca. How much fiber did Google own ? “ More than anyone else on the planet. ”

But there was only so much that could be done when someone else owned the facility. Google’s engineers knew that if they had a chance to design their facilities from the ground up — beginning with the site selection — they could be much more efficient. By mid - 2003, Google reluctantly began planning to build its own data centers.

There was a lot of unheeded literature about how to cool computers. One paper outlined a potential back - to - back arrangement of servers where the exhaust pipes faced each other and created a warm aisle between the racks. These would alternate with cool aisles, where the intakes on the front of the servers would draw on cooler air. Google tried to implement such arrangements in its colos, but the facilities managers complained.

Its own data centers would have enclosed rooms that segregated the hot air. Inside those separate rooms, the temperature would be much higher — perhaps 120 degrees or even more. If someone had to go into one of those hot rooms, you could temporarily cool the area down so the person wouldn’t melt while trying to swap out a motherboard. Even in the cold aisles, Google would raise the temperature. “ You can save just 20 percent by raising the thermostat, ” says Hölzle. “ Instead of setting the cold aisle temperature to 68 you can raise it to 80. ” Doing so would put a lot of stress on the equipment, but Google’s attitude was, so what if stuff broke ? “ You counted on failure, ” says Chris Sacca. “ We were buying nonspec parts [ components rejected for commercial use because they were not rated to perform at high standards ], so we didn’t need to coddle them. ”

Ultimately Google hired a small East Coast company called DLB Associates. “ I think they were actually not convinced at all in the beginning, but they were willing to collaborate, ” says Hölzle. So Google, a company that had once focused entirely on building Internet software, prepared to begin a building program that would lead it to construct more than a dozen billion - dollar facilities over the next few years.

A key Googler in the process was Chris Sacca. Not quite thirty at the time, Sacca had already been through several careers. Born to a working - class family in Buffalo, he’d been a ski bum, a stock speculator, and a lawyer. During law school at Georgetown, he’d taken a three - month break to help El Salvador’s telecom privatization project and after that thought he could make extra money as a consultant. ( To buff up his status, he gave himself a fancy name : The Salinger Group. ) When his stock trading put him $ 4 million in the hole, he took a job at a Silicon Valley law firm, where insistent networking put him on Google’s radar. “ They wanted one person who could identify, negotiate, draft, and close data centers, ” he says. “ My job was to get in the car, get on a plane, go find data centers to buy, lease.

In any case, when seeking out locations for a Google data center, Sacca and his colleagues did not let on who employed them. Sacca frequently used the name of his made - up consultancy operation, The Salinger Group. Other times he’d say he was from Hoya Technologies. 

The basic requirements for a data center were clear : land, power, and water. The last was important because the cooling process was to be done by an evaporative process that required millions of gallons of water through refrigerator - style “ chillers ” that drop the temperature and then run the cool water through “ jackets ” that hug the server racks. Then the water — heated up by now — gets run through massive cooling towers, where it trickles down, evaporates, and gets collected back in the system again. ( The air - conditioning is generally reserved for backup. ) All of this requires massive power, and before a shovel can be stuck into the ground, it has to be determined whether the local electric utility can provide sufficient amps to power a small city — at bargain rates.

The local congressman set up a conference call and mediated between Google and the Bonneville Power Administration. Then Google worked with the state to get fifteen years of tax relief, only the second time in Oregon history that a company had received a break of that length.     

After its construction, the building dominated the landscape, a massive shell the size of two football fields with a pair of four-story cooling towers. According to Sacca, the shell cost about $50 million, but its contents were valued close to a billion dollars. There was more than 200,000 square feet of space for the servers and infrastructure and another 18,000-square-foot building for cooling towers. In addition, there was a 20,000-square-foot administration building that included a Google-esque cafeteria and a dormitory-style building almost as large for transient workers. The exterior gave no clue about its contents.

“It didn’t matter whether you have 500 or 500,000 computers—you could run them remotely. We designed it for scale. We need physical hands only to get computers in place and replace the hard drives and motherboards when they fail. Even at the point where we had 50,000 computers, there were maybe six of us maintaining them.”          

“We could lose an entire data center, and everything would just spill over to the other data centers and we’d still have excess capacity,” says Jim Reese.

The data center in Saint-Ghislain, completed in 2008, actually eliminated chillers entirely. The average summer temperature in Brussels is around 70 degrees, but during the occasional scorcher, Google would shift the information load to other data centers.

Organizing Google’s hundreds of thousands of computers was one of those “hard problems” that make PhDs want to work at Google. It was definitely the lure for Luiz Barroso. He had been yet another colleague of Jeff Dean and Sanjay Ghemawat at Digital Equipment Corporation’s Western Research Lab.

But because of his hardware expertise, a couple of years after he arrived, Urs Hölzle asked him to help design Google’s data centers.

Indeed, a 2009 publication by Barroso and Urs Hölzle that described Google’s approach (without giving away too many of the family jewels) was called The Datacenter as a Computer.

By perfecting its software, owning its own fiber, and innovating in conservation techniques, Google was able to run its computers spending only a third of what its competitors paid.

One reason Sanjay Ghemawat loved Google was that when researchers were looking to solve problems a year out, Larry Page demanded that they work on problems that might be a decade out, or maybe even a problem that would come up only in a science fiction novel. Page’s point of view seemed to be, If you are ridiculously premature, how can people catch up to you?

The person in charge of Google’s strategy was the person who had first come up with the company’s business plan, and had been later instrumental in AdWords, the product that would make virtually all of Google’s money: Salar Kamangar.

After the deal closed in March 2006, the Writely team began migrating its product to the Google code base. The product became part of a project code-named Tricks, a web-based alternative to Microsoft Office. Google had already started developing a web-based spreadsheet that would be a companion to the word processor.

Google’s chief economist, Hal Varian, would later explain how it worked when new data centers open: “We’ll build a nice new data center and say, ‘Hey, Google Docs, would you move your machines over here?’ And they say, ‘Sure, next month.’ Because nobody wants to go through the disruption of shifting. So I suggested we run an auction similar to what airlines do when they oversell a plane—they keep offering bigger vouchers until enough customers are willing to give up their seats. In our case, we offer more machines in exchange for moving. One group might do it for fifty new ones, another for a hundred, and another won’t move unless we give them three hundred. So we give them to the lowest bidder—they get their extra capacity, and we get computation shifted to the new data center.”                

Google eventually devised an elaborate auction model for divvying up existing resources. In a paper entitled “Using a Market Economy to Provision Computer Resources Across Planet-wide Clusters,” a group of Google engineers, along with a Stanford professor of management science and engineering, reported a project that essentially made Google’s computational resources into a silicon Wall Street. Supply and demand worked here not to fix stock prices but to place a value on resources.

By the time it bought Zenter, Google had already released a beta version of its web-based productivity suite, Google Docs. Google Docs had one huge advantage over Microsoft Office: it was free. Google also began marketing a version to corporations, universities, government agencies, charging $50 a year “per seat” (i.e., for each user) for a license.

The surest sign that Schillace was right? In 2010, Microsoft rolled out an online version of its Office product—for free. Even if only a small percentage of the marketplace used Google’s own productivity apps, the company had achieved its larger goal—moving work onto the web.

In 2001, Page and Brin had told Schmidt they wanted Google to build its own browser. Right away.

Chan realized that users were ignoring the Toolbar because it provided no value to them. His idea was to implement a feature that would allow people to block annoying pop-up windows, which at the time were a plague on the net. But when he presented the idea at a meeting, Brin and Page, who had tied water bottles to the venetian blind cords and were playing a game of water-bottle tetherball, nixed the idea. “That’s the dumbest thing I’ve ever heard!” said Page. “Where did we find you?” Chan built the pop-up blocker anyway, and surreptitiously installed it on Page’s computer. (“He’d leave the computer on in his office,” says Chan.) Not long afterward, Page remarked that his browser was running faster. Chan told him that he’d installed the pop-up blocker. “Didn’t I tell you not to do that?” asked Page.  “Oh, it was a 20 percent project,” said Chan. Page dropped his suspicions and okayed the feature, which helped spur millions of Toolbar downloads.

The head of the Product Client Group was an intense engineer named Sundar Pichai. Born in Madras, India, he was among many Googlers who had attended the Indian Institute of Technology. After graduation, he followed the well-trod path to the United States and earned an M.S. in computer science at Stanford. But Pichai left academia in 1995. “The PhD just seemed like too long a commitment,” he says. “I just wanted to work.” He took various jobs in semiconductors and came to enjoy product management and business management, so he went to business school. 

Page and Brin wanted Chrome optimized to run web applications—fast. When you run a program faster by an order of magnitude, you haven’t made something better—you’ve made something new.

Sundar Pichai and his team had an OKR of 20 million users by the end of the year. “It was a very aggressive OKR,” he says. “A classic.” He didn’t make it. “We got there, but not in the time frame we had in mind.” A lot of people downloaded Chrome in the early weeks and found that it didn’t work. Because the online behavior of Googlers was not typical of the general public, there had been many websites and apps that went untested. “We had five thousand internal users, but not one noticed that Hotmail didn’t work,” says one engineer. But after Chrome shipped to the public, Hotmail users instantly found that it wouldn’t run their mail—and deleted Google’s browser.

Pichai gave a netbook to his father and noticed that once his dad opened Chrome, he never opened another application. He came to think that the word “application” didn’t apply to a browser—it was more like a gateway to everything in the world that really mattered, the stuff in the cloud. “It was very clear to us a lot of people were buying these devices with the goal of spending their entire day in the browser. So we all started talking about a natural course: designing an end-to-end experience around the browser. Think about it.”

They put together a proposal that they took to a meeting with Larry and Sergey in October 2008. Since Brin and Page had been wanting to do an OS for ten years, they instantly embraced the idea. “I’m all for it,” said Page.

“Think about it,” said Pichai. “You just want to get information into the cloud. When people use our Google Docs, there are no more files. You just start editing in the cloud, and there’s never a file.”


Part Five: Outside the Box: The Google Phone Company and the Google TV Company

Rubin, who was a maniacal robot aficionado—he would haunt the Akihabara district of Tokyo for weird Japanese toys, and build a few of his own—called the company Android. He gathered a team of eight to begin working on a prototype.

He would later vividly recall the trip he had made to Korea—“on my own dime!” he said—to present the concept to Samsung. He and two colleagues found themselves in a huge boardroom. Standing along the wall were about twenty carefully manicured executives in blue suits. (Rubin was in blue jeans.) The division head arrived, and, as if on cue, everyone sat down. Rubin gave his presentation, and the division head rocked with laughter. “You have eight people in your company,” said this executive. “And I have two thousand people working on something that’s not as ambitious.” It wasn’t a compliment.

Page had an idea: what if Google bought Android? It was a classic Larry Page moment: ask him to consider a toothpick, and right away, he was thinking about a forest. Later, Page would explain that he and Sergey had been thinking about getting deeper into mobile for a while. “We had that vision,” he says. “And Andy came along and we were like, ‘Yeah, we should do it. He’s the guy.’”

An open network would give Google unlimited opportunity, so that even if Google spent millions of dollars to develop an operating system—and then gave it away for free—it would still come out ahead.

So in July 2005, Android went to Google. The biggest adjustment Rubin had to make was keeping his limited-edition German sports car in the garage—in 2005, ostentation was still discouraged in the Google parking lots.

Rubin appreciated the interest of his bosses, but he appreciated even more the unusual degree of autonomy he was granted. He managed to get a special concession for his Android team: he could do his own hiring.

Jobs bonded especially with Brin; both lived in Palo Alto, and the pair would take long walks around the town and up in the hills … current and future kings of the Valley, inventing the future.

August 2006, Jobs invited Eric Schmidt to sit on Apple’s board of directors, which included Google board member Arthur Levinson, CEO of Genetech; and Bill Campbell, Google’s corporate coach. Al Gore sat on Apple’s board, while he was the self-described “virtual advisory board” at Google. Intel CEO Paul Otellini, who was on Google’s board, had started supplying the chips for Macintosh computers. There was so much overlap that it was almost as if Apple and Google were a single company.

Apparently, it took a while for Jobs to understand that Google was becoming his competitor. It was almost a year after the Android acquisition that Schmidt joined Apple’s board. “I feel I fully disclosed it when I joined,” says Schmidt, who adds that he also informed Jobs about the impending Chrome browser. But at that time, Jobs apparently believed that Google’s phone plans rested with the Sooner version, which was more of a competitor to Microsoft’s Windows Mobile than an iPhone rival. When Apple introduced its iPhone in January 2007, Jobs didn’t seem to be worried about Android, at least judging by the mutual good feeling when Jobs called Schmidt to the stage at the product launch. Schmidt joked that the collaboration was so close that the two firms might as well combine. “If we merge the companies we can call it AppleGoo … but we can merge without merging.”

Nonetheless, insiders say that over a period of months, Jobs concluded that he was a victim of deceit. The first alarming sign of Google pursuing its vision regardless of its effect on Apple was the Chrome browser. It competed with Apple’s Safari browser and also with the open-source WebKit technology that Apple had developed for Safari. It was all kosher from both a legal and an industry practices standpoint, but Jobs wasn’t happy, especially since Google had tried to hire some of his Safari developers. Android, though, was much worse. As he learned more about how the benign competitor he had envisioned was actually a full-blown alternative to the iPhone, Jobs became increasingly upset. Yet for months he was reluctant to break with Google. 

After increasingly contentious phone calls, in the summer of 2008, Jobs ventured to Mountain View to see the Android phone and personally judge the extent of the violation. He was reportedly furious. 

Jobs believed that Apple’s patents gave it exclusive rights to certain on-screen gestures—the pinch and the swipe, for example. 

In a blog post, Rubin stipulated, “We’re not announcing a Gphone.” Instead, he promised something more significant: a “truly-open and comprehensive platform for mobile devices … without the proprietary obstacles that have hindered mobile innovation.”

Whitt’s suggestion, Google became an active proponent for open spectrum. His team wrote up four conditions that the FCC should impose on whoever won the spectrum in the most valuable block of frequencies up for auction.

Developers even began to think of the Android operating system as a platform for bigger devices such as tablet computers or even small laptops. Barnes & Noble designed its Nook e-book reader around Android, and Asian manufacturers began making plans for Android-based netbooks.

Perhaps the breakthrough Android device came about a year after the original, when Verizon introduced an Android handset made by Motorola called the Droid. (This was a significant partnership since Verizon had become a fierce opponent of Google on technology policy issues. The thaw in relations was probably attributable to Verizon’s need to market a competitor to AT&T’s iPhone.)

“We don’t monetize the thing we create,” Andy Rubin says. “We monetize the people that use it. The more people that use our products, the more opportunity we have to advertise to them.”

At one point in 2003 Griffin and Sheryl Sandberg went to ask Larry Page for more people. He told them that the whole idea of support was ridiculous. “Why do we even do it?” he asked. Instead of Google assuming the nonscalable task of answering users one by one, it should let users help one another!

Larry Page laid out his own vision: Google Voice should become an Android application that would do everything that GrandCentral did—but it would also make Internet voice calls, like Skype. You wouldn’t even know it was any different from making regular phone calls. The plan would essentially make Google into a stealth phone carrier. *(Chan liked the idea—he thought that Page was brilliant on innovation but not so much on the details of product design. “Larry’s the worst person you want designing your product—he’s very smart but not your average user,” he says. To avoid this situation, Chan had a strategy of “giving him shiny objects to play with.” At the beginning of one Google Voice product review, for instance, he offered Page, and Brin as well, the opportunity to pick their own phone numbers for the new service. For the next hour the founders brainstormed sequences that embodied mathematical puns, while the product sailed through the review.)*

Halfway through the development cycle, an opportunity arose that Google’s leaders felt compelled to consider: Skype was available. It was a onetime chance to grab hundreds of millions of Internet voice customers, merging them with Google Voice to create an instant powerhouse. Wesley Chan believed that this was a bad move. Skype relied on a technology called peer to peer, which moved information cheaply and quickly through a decentralized network that emerged through the connections of users. But Google didn’t need that system because it had its own efficient infrastructure.

Chan was desperate to stop the acquisition, so he went to his friend Salar Kamangar. By then Kamangar, though almost totally unknown to the outside world, had become a hugely influential force in the company. He was a key member of a quiet cabal of Googlers who weighed in on crucial issues and influenced the final decisions of Brin and Page. Some were top executives, and others had “influence beyond their title,” says one insider, who said that quite often the conflicts aired in GPS meetings were settled by conversations and email among this loose cabal. The group included some of the very early people, such as Susan Wojcicki, Marissa Mayer, and  Lori Park, who had been one of the first twenty employees and had been influential in activities such as protecting the logs, China policy, and fulfilling odd personal assignments for Larry and Sergey. It wasn’t a formal club, and you didn’t have to have been at Google from the very early days—sometimes people like Chan could work their way in by being supersmart in a very Googley way, such as coming up with great ideas that promoted the company in general. Everyone in that inner circle really cared about Google both as a company and as a concept. Of all the people in that cohort, none was as respected as Kamangar. “Salar is like the secret president of Google,” says Chan, who laid out the reasons why a Skype acquisition would be a disaster. Kamangar agreed. Then the two of them talked to Sergey and won him over as well. With those allies on board, Chan devised a plan to kill the Skype purchase. As he later described it, his scheme involved “laying grenades” at the executive meeting where the purchase was up for approval. Chan tricked the business development executive who was pushing the acquisition into thinking that he was in favor of the deal: he had even prepared a PowerPoint presentation with all the reasons Google should buy Skype. Chan says that halfway through the presentation, though, the trap sprang. Brin suddenly began asking questions that the deck didn’t address. “Who’s going to run this?” he demanded. “Not me,” said Kamangar. Craig Walker said he had two kids in school and wasn’t about to make regular runs to Eastern Europe. “What are the regulatory risks?” A lawyer said it might take months to get approval. Finally, Brin looked at Chan and asked why Google would want to take the risk to begin with. Chan dropped his defense entirely and began explaining why Google had no need for Skype. “At that point,” recalls Chan, “Sergey gets up and says, ‘This is the dumbest shit I’ve ever seen.’ And Eric gets up and walks out of the room. The deal’s off.”

So Google bought it, changed the business model from a $1,000-a-year subscription to free, and integrated it into its Google Maps application—and into its mirror world. By 2009, 300 million people routinely peered down on the earth from space via Google Earth.

Larry asks himself in situations like this is not How can I help this person? Instead, he’s asking himself, Ten years from now, what thing can we build at scale that’s going to have the maximum impact on humanity?)

But none of Google’s postsearch initiatives would be more significant and have a bigger impact on the world at large—and draw a bigger lawsuit—than its movement into online video. The key was a 2006 corporate acquisition, the biggest the company had made to date, of a company whose name would become almost synonymous with the explosion of Internet video, a brand almost as recognizable as Google itself: YouTube.

In February 2005, Karim, Chen, and Hurley formed a company called YouTube. (Karim, who wanted to return to academia, soon went back to school and turned over the leadership to his partners.) They set up shop in a second-floor office over a pizzeria in San Mateo, halfway between San Francisco and Palo Alto.

Even though YouTubers knew that people who were uploading videos didn’t really have the right to do so, they believed that YouTube would be all right as long as there weren’t complaints from copyright holders about specific videos, in which case they could respond. Otherwise, they would simply assume that copyright holders permitted their content to appear on YouTube. Chen’s instinct in this case turned out to be a canny interpretation of the Digital Millennium Copyright Act, which promised a “safe harbor” to sites hosting uploaded content. But the decision to take a lax view of policing copyright was less likely a legal judgment than one determined by the carpe diem ethic of a start-up.

In contrast, YouTube was dead simple: everything was free, you could find clips from just about anything, and it played inside your browser. God knows where its users had gotten access to some of the stuff they put up there, but because of the company’s lax policy of policing its archives, YouTube managed to have just about anything you were looking for. YouTube users had uploaded a popular clip from Saturday Night Live called “Lazy Sunday,” which became a phenomenon—5 million people streamed it until NBC demanded that YouTube remove the clip seven weeks after its appearance. The clip jacked up YouTube’s traffic by 83 percent. Later, it was cited as the event that restored luster to the aging SNL.

The Google Video team acknowledged that “the user-generated trend is huge” but didn’t seem to grasp how dominant YouTube was becoming—the little start-up located over a pizza shop was streaming 25 million videos every day, more than three times as many as Google. The Google Video team seemed to take comfort in reporting that premium content owners—which it still considered the key players in the field—viewed YouTube as “a small start-up with no cash,” “perceived as trafficking in mostly illegal content.”

To build up infrastructure would require a lot more money than the original $3.5 million in venture capital it had received from Sequoia. YouTube got another round of funding for a total of $11.5 million, but even then it would struggle. Serving millions of videos a day was just plain expensive.  

Schmidt would later say in a deposition in the Viacom lawsuit that he estimated that YouTube’s worth at that point was between $600 million and $700 million. “It’s just my judgment,” he said. “I’ve been doing this a long time.” But Google wound up paying $1.65 billion to close the deal with YouTube. “I’m not very good at math,” said the deposing attorney, “but I think that would be $1 billion or so more than you thought the company was, in fact, worth.” Schmidt provided an excellent summary of deal making in Internet time, embodying the Google principles of speed, scale, and minimizing opportunity cost. This is a company with very little revenue, growing quickly with user adoption, growing much faster than Google Video, which is the product that Google had…. In the deal dynamics, the price, remember, is not set by my judgment or financial model, or discounted cash flow. It’s set by what people are willing to pay. And we ultimately concluded that $1.65 billion included a premium for moving quickly and making sure we could participate in the user success in YouTube.

 If Google had been inclined toward remorse about the price, such worries were surely mitigated by a letter sent by Rupert Murdoch’s Twentieth Century Fox as the deal was closing. It declared that whatever Google was paying, Fox would pay more.

Brin’s wife, Anne Wojcicki, was a cofounder of 23andMe, a company involved in personal DNA analysis. Brin defied corporate propriety when he shifted his personal investment in the firm to a company one. Google’s lawyers made sure the transaction passed formal muster.

Brin would subsequently attempt to stave off the onset of Parkinson’s with a self-determined regimen of physical activity—he took up diving—and by imbibing gallons of green tea. “This is all off the cuff,” he told Wired reporter Thomas Goetz, “but let’s say based on diet, exercise, and so forth, I can get my risk down by half, to about 25 percent.”

Page married Stanford graduate Lucinda “Lucy” Southworth and worked even harder to keep his personal life out of public view. It was a life much different from the modest one of a grad school dropout that he had led for the first few years of Google. He held his wedding free from web snoops on the isolated Caribbean island owned by a fellow billionaire, Richard Branson, the British head of the Virgin group of companies.


Part Six: GuGe: Google’s Moral Dilemma in China

“How many engineers does Microsoft have?” asked Page. About 25,000, Page was told. “We should have a million,” said Page. Eric, accustomed to Page’s hyperbolic responses by then, said, “Come on, Larry, let’s be real.” But Page had a real vision: just as Google’s hardware would be spread around the world in hundreds of thousands of server racks, Google’s brainpower would be similarly dispersed, revolutionizing the spread of information while speaking the local language.


Different countries presented different challenges. In India the politicians demanded penalties and censorship when users of the Orkut social-networking service, very popular in that country, launched epithets at officials. In Thailand, the king could not be insulted. In Germany, denying the Holocaust is illegal. Generally, in cases where officials ordered that Google filter its search results, the company would push back. It was a constant struggle. But nothing like that in China.

This ascent came to an abrupt halt on September 3, 2002. That day, Chinese visitors who typed “” into their browsers got only error messages. The Great Chinese Firewall had blocked Google. That was how outsiders referred to the technology behind the Chinese government’s sweeping censorship. China realized that the Internet was a commercial necessity but the potential freedom of speech it offered was deemed a threat. So the country built an elaborate censorship infrastructure to block disfavored sites or pages.


Lee resigned from Microsoft on July 18 and officially accepted Google’s offer the next day. It was worth over $13 million, including a $2.5 million signing bonus. Lee posted an explanation on his Chinese-language website, with the headline “I need to follow my heart.” He said that Google had given him “a shock” by its fresh approach to technology and postulated that in China, his new employer’s youth, freedom, transparency, and honesty would produce a miracle. “I have the right to make my choice,” he wrote. “I choose Google. I choose China.”

Ironically, in early 2002, Kai-Fu Lee, who was an early enthusiast of Google search, had once recommended to Bill Gates that Microsoft buy Google. After looking into it, Gates told Lee that the cost would be too high.

Despite Microsoft’s saber rattling, Lee would get his chance to work with Google. On September 13, Judge Steven Gonzalez ruled that while Lee was prohibited from sharing proprietary information with or helping Google in competitive areas such as search and speech technologies, he could participate in planning and recruiting for Google’s effort in China. Ultimately, the two companies would settle, and the restrictions on Lee’s activities would be lifted in 2006.

According to Chan, at one meeting a number of people Lee hired in China began squabbling about what their titles should be. “Your title,” Chan told them, “is product manager.” They objected that in China no one knew what that meant, and they preferred the official appellation of “special assistant to Kai-Fu Lee,” so everyone would know that they had the ear of the esteemed leader of Google China. Chan almost fell over. “This isn’t the White House!” he told them. “Our job is to be focused on users, not Kai-Fu.” But they insisted and told him that it was important for them to sit within a hundred feet of Lee, a geographic honor that would cement their status as special assistants. Worst of all, he says, “It was this weird culture of kiss up or kiss down, and I really don’t do that. So I said, ‘Okay, I’m done.’” Besides, the air pollution in Beijing was killing him.

Baidu ruled search in China. It was founded by Robin Li—the Chinese native who had discovered the power of web links in Internet search at the same time that Larry Page and Jon Kleinberg had. He had left the United States in 2000. “I didn’t have a Stanford degree, and I didn’t know many VCs at the time, so I went back to China and started to develop our own search technology,” he says. (Despite this, his new company was funded by Silicon Valley VC money.)


But it also took advantage of a freedom that Google did not have, particularly in flouting copyright regulations. A significant percentage of its searches were for music, and the links that came up on results connected users directly to free downloads of songs. It was such a dominant music distribution tool that Chinese people call MP3 players “Baidu devices.”


The Baidu founder was particularly unimpressed with Kai-Fu Lee’s recruiting strategy. “I had been very afraid of Google hiring away my engineers by doubling or tripling their salaries. Instead, they hired a lot of fresh graduates and brought Chinese engineers from Mountain View to train them. That gave me some relief,” he says.

Baidu had succeeded in transforming the competition into a test of patriotism. Its message was that Baidu, being local, understood China and Google did not.

But Baidu’s biggest boost came from the Chinese government. The government would often slow or block Google’s service and at one point even redirected Google traffic to Baidu.


But Mark Zuckerberg was in the Larry Page mold, a wildly ambitious leader with a quasi-religious trust in engineering. Zuckerberg said that Facebook would have hacker values. Ten years younger than Page and Brin—a generation in Internet time—Zuckerberg respected Google’s values but believed that the older company had lost its nimbleness and focus. He made a specialty of hiring Google people who sought the excitement of building something new. When Zuckerberg needed a strong number two to run Facebook operations, he turned to Sheryl Sandberg, who had built Google’s ad organization.


The irony was that Google had been present at the explosion of social networking. In a classic case of the company’s corporate ADD, it simply had blown the opportunity to make the most of what it created. Back in 2002, a young Google engineer named Orkut Buyukkokten had an idea. “My dream was to connect all the Internet users so they can relate to each other,” he later recalled. “It can make such a difference in people’s lives.” Buyukkokten, who had come to Google from his native Turkey via Stanford, decided to use his 20 percent discretionary time developing a cyberspace preserve where the people of the world could intermingle in peace, presumably so their good vibes would go viral.


Not long afterward, Crowley saw the first version of an Internet start-up called Twitter.


Crowley remembers a fateful videoconference with Mountain View in the summer of 2006 in which he and his colleagues argued that the social network movement was about to go crazy and now was the time to put more resources into Dodgeball. An executive flatly told him once and for all to forget about asking for more engineers. That sealed it for Crowley. Though reluctant to abandon Dodgeball’s loyal community, he and Rainert left in April 2007.

In theory, Twitter was so simple that Google could simply write its own version. “The question of the day was ‘Why don’t we build Twitter?’ Three guys could do it in a weekend!” said Glazer in 2009. But, he explained, that would have been a case of chasing taillights.

It was an indication of Google’s confused strategy that Taco Town’s development proceeded even as the company announced Wave with fanfare and hosannas.


As the team prepared for Taco Town’s rollout in early 2010, the product added more features, many of which duplicated Facebook functions. It also added location information to the Tacos. But the minipostings would no longer be called Tacos; Google renamed the product Buzz to reflect the crackling interaction it would presumably generate.

But within forty-eight hours, Buzz ignited a privacy crisis as intense as the Gmail privacy conflagration in 2004.


Previously, new users of social networking services had been confronted with the annoying chore of gathering friends and contacts to construct their cohort. Google felt it had solved this problem with Buzz.


But when the general public tried Buzz, some users discovered unwanted—even horrendous—consequences. By looking at a Buzz user’s profile, other Buzz users could see that person’s social network. Since the network had not been carefully built contact by contact, it was entirely possible that it might include a connection that a user might not want exposed to a larger audience. (Certain contacts could indicate that someone was seeking alternative employment or spilling secrets to a reporter.) As described in a February 10 posting in Business Insider, “The problem is that—by default—the people you follow and the people who follow you are made public to anyone who looks at your profile … someone could go in your profile and see the people you email with and chat with most.”

Google had made a critical error. Its employees differed from the general population. For one thing, their email contacts are largely with other Googlers.

That March, Urs Hölzle sounded an alarum that evoked Bill Gates’s legendary 1995 “Internet Sea Change” missive to his minions at Microsoft. Just as the Internet threatened Microsoft back then, in 2010 the sea change to a more people-oriented Internet—social media—was becoming a problem for Google. Hölzle said that the challenge required a decisive and substantial response, involving a significant deployment of personnel—right away. The memo became known as the Urs-Quake.

“Day-to-day adult supervision no longer required,” Schmidt tweeted. The veracity of that statement remains to be seen. But one thing seems indisputable: Larry Page would not be a conventional CEO. Google’s future would continue to court the unexpected. And maybe the impossible.



Popular posts from this blog

Foundational distributed systems papers

Your attitude determines your success

Progress beats perfect

Cores that don't count

Silent data corruptions at scale

Learning about distributed systems: where to start?

Read papers, Not too much, Mostly foundational ones

Sundial: Fault-tolerant Clock Synchronization for Datacenters


Using Lightweight Formal Methods to Validate a Key-Value Storage Node in Amazon S3 (SOSP21)