Episode 1: A History of Chinese Entrepreneurship in Three Biographies
In September 2017 a Bloomberg News headline got my attention. “This Startup Is Luring Top Talent With $3 Million Pay Packages.” To me, a recruiter in Silicon Valley, that’s about as good as clickbait gets. I’m used to speaking with software engineers from established companies like Google, Apple, and Facebook, who expect skyscraping compensation. But $3 million was suspiciously high. Especially for a startup.
The article was about a Beijing-based company called ByteDance that was behind a popular mobile application called JinRi TouTiao, or “Today’s Headlines.”
Just about a month later, another story broke: ByteDance had tried to buy Reddit, a website which I am reluctant to admit that I use almost daily for a dose of dog pictures and dopamine.
In the span of two months, ByteDance had gone from a company that I had never heard of before, to bidding on America’s fourth most popular website.
I was blindsided. And I knew that if I was blindsided, the rest of Silicon Valley was, too. I speak Chinese and lived in China for three years before returning to the US in the Fall of 2014. I even did an internship during that time with a Chinese startup called Wandou Labs, working on their now defunct international product, SnapPea.
So what the hell was going on? What was ByteDance? And how had I––a Mandarin-speaking tech industry worker––missed its rise into a major player in the tech industry with a $20 billion dollar valuation and the audacious confidence to bid one of my favorite websites?
The short answer is that I blinked.
But today’s episode is about the long answer. I am going to tell three stories about Chinese technology entrepreneurs. With these stories, I argue that we have entered a new era of innovation in China. Whereas China used to rely on the United States, the West, and the English language for innovation, today it does not. And as a result, anyone in the technology industry who is not paying attention to China, Chinese entrepreneurs, and Chinese companies, will not just be repeatedly blindsided, but are effectively blind.
Part 1: Tsinghua to MIT
I want to start by introducing you to a Chinese American immigrant named Shih Ying Lee, who is best known for being Professor Emeritus at M.I.T. as well as the founder of a Boston manufacturer of industrial sensors called Setra Systems. Lee is a remarkable man. In addition to his professional accomplishments, he played tennis into his late 80s, published an autobiography at age 89 in 2007, and this year will celebrate his 100th birthday.
His story is an amazing illustration of what Chinese innovation looked like during the mid-20th century. I hope Lee’s story will give listeners a sense of just how much things have changed.
Lee was born in Beijing in 1918, a time when public education in China had existed for only a bit over a decade. His earliest years were spent learning Chinese classics with a private tutor.
When his parents finally did put him into the test-oriented Chinese public education system, he excelled. He passed one test to get into a top high school, and then another that won him a spot in Tsinghua University, which even one hundred years later is still China’s top engineering school.
Lee entered Tsinghua in 1936 as a physics major. It was an elegant campus, funded in part by the United States through an initiative of Teddy Roosevelt. For any listening graduates of private school, I’ll merely note that tuition at Tsinghua at the time was roughly 20 Chinese dollars per year. That’s $71.46––that’s US dollars––today.
In a pattern that would become familiar to Chinese students throughout the 20th century, Lee’s studies were interrupted by conflict. While Lee and a classmate enjoyed a summer vacation trip to the ocean town of Qingdao, the Second Sino-Japanese war began. Lee reluctantly admitted to himself that it would be too dangerous to try to return to his family in Beijing, and instead decided to flee. He would not return to Beijing until 1973.
With the Japanese occupying Tsinghua’s Beijing campus, the university opened up a temporary campus in Changsha, a city in Southern China, where Lee was able to continue his studies, switching his major to electrical engineering. His sophomore year was interrupted by the advancing Japanese army, and the university relocated again, this time to Kunming, a city southwest of Changsha. Being wartime, there were no vehicles to transport the students––so many of Lee’s classmates walked nearly 1,000 miles to the new campus. That should give you a sense of how serious Tsinghua students were back then––and perhaps still are today.
After three years in Kunming, Lee graduated and found a job doing research in a government hydraulics lab. Eventually, he received a scholarship to attend the University of Iowa, the same university that sitting Chinese President Xi Jinping attended in 1985.
Lee flew from China to India and then sailed to the United States aboard the S.S. Mariposa which had brought American troops to Asia as part of the war effort.
Once stateside, Lee decided to apply to M.I.T. and was admitted. So he skipped out on De Moines and went to Boston instead. Sorry Hawkeyes... Can you really blame him?
Because most college-aged American men were serving in World War II, Lee found himself in classes with just five or six students. After just two-and-a-half years, Lee left M.I.T. with a Doctorate in Civil Engineering.
When Lee graduated in 1945, the Chinese Exclusion Act––which banned Chinese from immigrating to the United States––had been repealed just two years earlier, which meant Lee had a path to immigration.
It was rare for Chinese Americans at the time to be in industry. Lee notes in his autobiography that Chinese either worked in universities or in Chinatown. But Lee found a job working as an engineer on Boston’s John Hancock building with the firm Cram and Ferguson––which is still in business today.
A year later, Lee returned to MIT to work in a laboratory which was developing a flight simulator for airplanes. Good work eventually earned him an assistant professorship teaching fluid mechanics to undergraduates.
Lee’s older brother, Y.T., another MIT graduate, had served in the Chinese airforce during World War II, but returned to MIT after the war to do a post-doc in Aeronautic Engineering. In 1951, the brothers began to collaborate. Y.T. pioneered an instrument that could measure the fuel flow in jet engines, while Lee developed a new control valve for use on airplanes and missiles. The brothers used royalties from the inventions as seed money for their first company: the Dynamic Instrument Corporation, later renamed Dynisco.
The brothers rented space near MIT and began to work on developing a new pressure sensor. It was not an especially capital-intensive business, but each sensor had to be hand-wound with wires. The business eventually grew to 70 employees when the brothers sold it in 1960.
As part of the acquisition, the brothers agreed to a non-compete for six years. But in 1967, they had a new idea. Lee had observed that there had been breakthroughs in the semiconductor industry. Basically, smaller and smaller electronics were now available. He believed that built-in electronics could be a game-changer in the pressure sensor industry, which he and his brother knew well.
That was the birth of Setra Systems. The Lee brothers set up a 400 square foot office and a machine shop in the lower level of the house. Not long afterward, they released their first product: an accelerometer. The product did well, selling to automotive and aerospace companies. The company grew and moved to an industrial park where they shared a building with another startup whose product you may be using this very moment: Bose Corporation.
The brother raised $400,000, which is about 2 and a half million today. With the capital injection, Setra was realizing consistent annual profits by 1976. Around the same time, Lee left his role at MIT and went full-time at Setra.
Setra eventually sold to Danaher in 2001, netting their original investors about 65 times their original investment. I want to note that Lee was still an active board member at the time, at 83 years young.
Lee is obviously an extraordinary individual, but in many ways his story is representative of how Chinese entrepreneurs innovated throughout the 20th century: A brilliant individual took advantage of government scholarships and family ties to find a way out of China, into higher education, and then into business in his new country.
Because of war, political turmoil, and backward economic policy, innovation couldn’t happen in China. Had Lee never left China––or had he returned to China after finishing at MIT––he may have spent his life as a researcher in a poorly funded Chinese government hydraulics lab, or even worse, ended up as a political victim of the Cultural Revolution. Instead, he stayed in America and had a truly remarkable career.
Staying in America was basically the only option for would-be Chinese entrepreneurs until the death of Mao Zedong, the end of the cultural revolution, and the 1980s reforms of Deng Xiaoping. You can see a story like Lee’s in the biographies of almost all 20th century Chinese innovators, such as An Wang, the founder of Wang Laboratories, and Ping Fu, founder of Geomagic and the author of the 2012 memoir Bend, Not Break.
But finally in the early 1990s, things started to change, and some brilliant minds could see opportunity arising in China.
Part 2: Advertise to 40 million people, 24 hours a day, 7 days a week, for $1/day, What do you think?
I’m going to turn to the story of another man named Li. A quick note, I thought about using someone else as an example, simply because I thought it might be inconvenient to use two individuals with the surname Li. But I ended up keeping it because it demonstrates something interesting: Over 90 million people in China have the surname Li. It’s the second most common surname in China. In case you’re wondering, the most common surname is Wang.
The second Li is Robin Li, the founder of Baidu, which is known to most Americans as “Chinese Google.”
Robin Li was born to factory workers in Yangquan City in Shanxi Province. Even today Yangquan is a small city by Chinese standards, with just over a million residents––the kind of city best known for being the birthplace of Robin Li as well as science fiction writer Cixin Liu, whose award-winning Three Body Problem may be made into an Amazon TV series. It’s a great read. I highly recommend it to sci-fi fans.
In 1987, Li received his city’s best grade on the college entrance exam and chose to head to Peking University to study information management. Li was a sophomore in 1989, when some of his classmates were protesting on Tiananmen Square. While Li was as apolitical, he shared a pessimism about China’s future that was common among young people at the time. He sensed that a career in China in information management would mean a dead end job as a bureaucrat, so he set out to study computer science in the United States.
He ended up at SUNY Buffalo, with a plan to get a PhD. But after receiving his Master’s and completing an internship in 1994 with a Dow Jones subsidiary called IDD Information Services in New Jersey, he decided to leave academia. Li took a full-time job at IDD, where he worked on the Wall Street Journal online edition, and stayed there until 1997.
Those three years were critical in the development of search engines. In 1996, two Stanford grad students, Larry Page and Sergey Brin, had developed an algorithm called PageRank, which used link analysis to determine the relevance of web pages. That algorithm, of course, was the seed that would become Google. Less well known, is that when Larry Page filed a patent application for Google PageRank, he cited a search engine called RankDex, which was the first search engine that used hyperlinks to measure the quality of websites it was indexing. RankDex was the work of an IDD Information Services engineer named Robin Li.
Li inched closer to entrepreneurship, taking a job with InfoSeek, a Silicon Valley startup search engine in 1997. Disney bought a 43% stake in the company in 1998, which most people in Silicon Valley will recognize as a sort of grim reaper for innovation.
I have to say, I think Li must have been frustrated with how things were going for him. Because he was clearly set on entrepreneurship from the beginning. Li was hustler.
Li wrote a few posts online during those years that you can see in the Usenet archives. Usenet was an early online forum that computer scientists and engineers at universities and technology companies used to talk to each other.
Here’s a post Li made on Usenet on December 8th, 1995––this would have been while he was at IDD in New Jersey:
Subject Line: Advertise to 40 million people, 24 hours a day, 7 days a week, for $1/day, What do you think?
Body text: Yes, I am talking about having your own WWW home page.
We design, host and market your home pages for you.
We have the best quality web server with ultra fast T3
connection to the Internet. Our professional web design team
developed the Wall Street Journal Interactive Edition.”
If you’re paying attention, that’s Li referring to the work he did with IDD.
Here’s another post, this one from July 8th, 1997:
“Subject Line: Patent on hypertext indexing and retrieval
Body text: I have a pending patent on hypertext indexing and retrieval which works fundamentally different from traditional search engines and is much better and faster. It can also be applied to automatically select "best of the web" in any specified areas.
I am looking for investors and partners on this idea.
Please reply to this account or call”––and then he lists a phone number with a New Jersey area code. I called it, just because it would be kind of cool if the CEO of Baidu picked up. But he didn’t.
I’ll merely note here that he was still signing his name as Yanhong Li, as late as 1998. He must have adopted the name Robin after Baidu took off.
One of the mysteries of Li’s story is why he didn’t start a company in the United States. He clearly had the technical chops to challenge Google. Was he afraid of losing his visa? Were investors blowing him off? Was he getting pressure from back home to stay in a corporate job? Or maybe he just wanted to go back to China the whole time––we do know that in 1995, he had begun to make annual trips back to China, feeling out the development of the internet there.
Whatever the reason, in 1999, he and his cofounder Eric Xu moved back to China and began work on Baidu.
The team rented two rooms in a three-star hotel near Peking University, and they got to work. Li negotiated with IDD for a license to use the RankDex algorithm––which must have been a bitter tasting pill for Li, paying to use his own algorithm. Xu took care of operational tasks, like finding office space:
Let me explain that laughter you hear. 1-4-1-4 in Chinese, is pronounced “Yao-si-Yao-Si.” But if you’re not careful about your pronunciation, it can sound like you’re saying, “I want to die, I want to die.” Which is funny, I guess, unless you’re a non-native Chinese speaker making the pronunciation mistake. Many office buildings in China skip the fourth and fourteenth floors.
Despite the inauspicious room numbers, the team had raised 1.2 million dollars at during the height of an internet bubble in China. 1998 and 1999 were unquestionably the best years to start an internet company in China: Tencent, JD, Sina, Ctrip, and Alibaba, five of China’s ten largest tech companies were all founded in either 1998 or 1999. Baidu was just a hair late to the party: its official founding date is January 1st, 2000.
Li made sure that his investors understood that the 1.2 million would be enough to sustain the Baidu team for six months, at which point, they would seek to raise more money. But Li cleverly told his team, that the money had to last them a full year. He was concerned that the bubble would burst. It did. And when it did, many companies went under. But Li’s frugal team at Baidu still had six months of runway, which they used to raise $10 million in difficult fundraising environment. Five years Baidu IPO’d.
Li and his cofounder Xu had both lived and worked in Silicon Valley. When they went back to China, they did their best to bring Silicon Valley with them.
Just after the team started working on Baidu, Li published a book with the Tsinghua University Press called “Silicon Valley Business Strategy.” It’s a series of stories about people like Mark Andreesen, Jim Clark, and Steve Jobs, and companies like Sun Microsystems and Microsoft.
At the same time, Li’s cofounder Xu produced a TV documentary called “Entering Silicon Valley.”
Here’s Xu again, talking about the subject:
And this is, I guess, the point I’m trying to make about these guys: They were students of Silicon Valley. As were an entire generation of Chinese entrepreneurs and innovators. Some of them, like Li and Xu, were educated overseas. While others, merely saw ideas that worked overseas and brought them to China––this is the generation, after all, that made “copycat China” a cliché. Even if sometimes, like in the case of RankDex, it’s not entirely clear who was doing the copying.
Part 3: Indigenous Innovation
When Baidu IPO’d Yiming Zhang had just graduated from Nankai University in Tianjin. Today, Zhang is 36, and the founder of ByteDance, the company I described at the beginning of this program. As of April 2018, Zhang still doesn’t have a WikiPedia page, which must be some kind of record, given his company’s $20 billion dollar valuation and a personal net worth estimated at $4 billion.
So let me fill in his bio for you: Zhang was born in 1983 in a city called Longyan in Fujian Province, which is a southern coastal province opposite from Taiwan. The 1980s were an exciting time for that region. The government had taken steps to liberalize the economy, and Fujian is close enough to Taiwan and Hong Kong, that Fujianese had a sense of what was possible, and perhaps even what was imminent.
Zhang’s parents often spoke about technology. His father worked for Longyan’s Science and Technology Commission before opening an electronics factory in the gritty industrial city of Dongguan near Hong Kong.
This all said, it makes sense that Zhang entered Nankai University as an electronics major, later switching to software engineering. After graduating, Zhang joined Microsoft, but left promptly, feeling that his work was aimless and far removed from the people who would actually be using the software.
He had posted his software engineering credentials and contact information to an online forum, and another Nankai alum contacted him about joining a travel industry startup. The startup was called Kuxun, which––I’d like to mention––is difficult to pronounce even if you use their English language name Kooxoo––K-O-O-X-O-O.
After a couple years at Kuxun, Zhang left for a startup microblogging platform called Fanfou, founded by another genius techie from Longyan City, named Xing Wang.
Things must have looked optimistic for the Fanfou team. Fanfou was the original “Twitter for China.” They beat everyone to market. And Wang, their CEO, was already proven. He had sold his first company in 2006.
But less than a year after Zhang joined Fanfou, the website was suddenly forced to shut down in July of 2009 in the wake of protests and subsequent deaths in Urumqi. Protesters had used online forums and chat groups to organize. The Chinese government censors were spooked and responded firmly. Remember that those were early days for online activism––the Arab Spring happened over a year later in December 2010.
Interestingly, Fanfou’s shut down left the door wide open for Sina Weibo, which was launched just two months afterward. Sina dominates microblogging to this day. That must have been incredibly frustrating for the Fanfou team, and I’m sure that Zhang, in his role as CEO of ByteDance, a media company, often remembers the experience.
Just a couple months after Fanfou shut down, TripAdvisor acquired Kuxun, which perhaps put some money into Zhang’s pocket. Because at the same time he decided to found his first company.
Before I go on, I wanted to tie up a loose end. Xing Wang, the Fanfou CEO, is now a billionaire in his own right. After Fanfou he founded Meituan, China’s dominant group buying company. As it happens, in 2015 Meituan acquired Kuxun from TripAdvisor, which brings this quite neatly full circle.
So Zhang started his first company. He raised money from SIG and founded 99Fang, a real estate search engine. Real estate is incredibly competitive in China, and it’s safe to say 99Fang didn’t go quite as well as the team had hoped. Today the website is defunct. But after a few years, Zhang and his team pivoted and in 2012 he founded ByteDance.
ByteDance’s flagship application is called Jinri Toutiao, or Today’s Headlines. It is an information content platform, most analogous, perhaps, to Reddit, but with a much greater emphasis on artificially intelligent content recommendations.
Whereas 99Fang topped out around 100,000 daily active users, just eighteen months after launching “Today’s Headlines,” the app had 10 million daily active users. As of September 2017, it had over 120 million.
In a future episode I might get into the gritty of what exactly makes Today’s Headlines so compelling. But the story of this episode is not the app, but the man behind the app.
Zhang didn’t study overseas. He snubbed work in an American company in favor of joining a Chinese startup. Unlike, Shi Ying Lee and Robin Li, I can’t find any video evidence that Zhang speaks English––though to be fair, as a graduate of Nankai, it’s likely that his English is at least as good as my Chinese.
And all this is important, because people like Zhang are the present and future of innovation in China.
The lesson from the progression of these stories––from the American immigrant Shih Ying Lee to the Western educated returnee Robin Lee, and finally to the homegrown Yiming Zhang––is that it’s going to get harder and harder to see innovation coming from China. Chinese technology entrepreneurs are going to be––in many ways––more and more distant from America and Silicon Valley.
And at the same time, it’s clear that China can innovate, and that Chinese companies and entrepreneurs are operating at the same cutting edge as their American counterparts. So I think it’s time for technologists to start paying closer attention to China.
Walter Isaacson’s biography of Steve Jobs was a best-seller in China. I know because I was living in China at the time and when I told people I was from Silicon Valley, everyone wanted to talk to me about Steve Jobs. But nobody has even bothered to translate Robin Li’s “Silicon Valley Business Strategy” into English.
Is that hubris? It might be hubris, but it also might be something else: If you’ve never been to China and you don’t speak Chinese––hell, even if you do speak Chinese––China, and its technology industry––can be really hard to approach. Where would you even begin?
That’s why I started this podcast: to try make tech in China just a little more approachable. So where should you begin? Right here.