Zhang’s childhood followed a relatively normal path in the northern Chinese city of Harbin until age 15, when the Cultural Revolution shuttered schools throughout the country. Over the next few years, city students were sent to the countryside for “re-education.” Zhang ended up spending almost seven years in the country.
“One thing I can say for certain is that my career path is different from anyone else,” she said in an interview. “I worked on the farm for five years as a tractor driver. Later I also worked in a rural area factory for another two years.”
It was an incongruous start for someone destined to play an important role in the development of the Internet. And yet, being a self-motivated student with a profound interest in science and an abiding curiosity about how things work, Zhang was attracted to Internet development in its early years and made seminal contributions to its architecture, a role that is reflected in her admission to the internet Hall of Fame.
Indeed, just a few years removed from her job in the relay factory in rural China, Zhang found herself in the United States – as an exchange student at California State University, Los Angeles.
“I got to Cal State and thought, ‘Wow, this is heaven,’” she said.
And a mere two years later, despite an atypical – and limited – formal education, Zhang was enrolled as a PhD student at MIT. From there, she went on to Xerox Park in Palo Alto and then back to LA for a long career as a distinguished faculty member at UCLA, where she remains today.
If Zhang is aware of her remarkable path and successes, she is quick to say that luck played a major role along the way, and as examples of serendipity she points to chance meetings with two benefactors who would change her life.
The first was an encounter with Fleur Yano, a physics professor at Cal State LA, who visited Harbin Institute of Technology shortly after Zhang had been admitted.
The second was bumping into Internet pioneer David Clark who became her adviser at MIT.
“I owe so much to them both,” Zhang said. “Without Professor Yano, who just happened to visit HIT when she was an acting dean, I wouldn’t have gotten into Cal State or even come to the U.S. And Dave willingly took me on as a PhD student. He supported me through my graduate school and taught me everything I know about the Internet.”
Luck, of course, has a tendency to bless those equipped to take advantage of it. Zhang might not have capitalized on these chance encounters had she not grown up with a powerful desire to learn.
“From a very young age, I wanted to go to university – not to study arts or language or history, but to study math and science and engineering,” she said. “All my life, I’ve always asked why – why things work the way they do.”
Beginning in elementary school, Zhang took to visiting a bookstore near her home, and she read avidly about science, math and technology. Her aptitude was such that her math teacher had her develop quizzes for the rest of the class during the morning self-study period.
And that intellectual hunger stayed with her. She never stopped reading, never stopped asking why, which paid off years later when schools began reopening throughout China.
Still on the farm, Zhang took an exam required to enter the university. And against all odds, given the long hiatus in her formal education, she succeeded and was admitted to Heilongjiang University in Harbin.
Zhang spent three years at Heilongjiang, but once again her education relied largely on her own initiative. Although the school was called a university, the Cultural Revolution had so gutted the educational system that Zhang’s classes were filled with students of widely divergent ages and abilities, including many who had yet even to enter middle school. Instruction tended to be basic, and politics still permeated teaching.
“We were told, ‘You need to get back to the workers and farmers to get educated.’ And for three years in that university, we didn’t learn much but we were constantly sent out to work as laborers,” she said.
Even this unorthodox bit of schooling came to an end in 1976, and Zhang returned to the countryside for two more years, this time in the electrical relay factory.
But in late 1976, the Cultural Revolution was winding down, and Zhang accelerated her self-study. In 1978, she successfully passed the entrance exam to the graduate program at Harbin Institute of Technology.
“I did better on the test than most, and I think it’s because I had always studied on my own,” she said. “There’s an old saying that ‘Among dwarves, you pick the tallest, even if it’s only by a.millimeter.’ I just got that millimeter.”
This time Zhang was actually able to study academics. She worked hard and was doing well when a first major stroke of luck occurred in the form of a campus visitor.
That person was Yano. Per a request from HIT, Yano agreed to help get four students enrolled in Cal State LA, one of whom was Zhang.
Although lacking the education credentials for admission to Cal State’s science programs, Zhang was accepted in 1979 in the school’s TESL program for teaching English as a second language and, once on campus, it wasn’t long before she was officially admitted to Cal State LA’s masters program studying computer engineering.
Coming to an American university was a revelation.
“I requested a small cubicle in the school library, and I was there all day and night until it closed,” she said. “In China, books were scarce, and it was hard to find interesting books other than on politics. Here was a whole library – it felt like the best in the world.”
Two books in particular stood out: Andrew S. Tanenbaum’s “Computer Networks” and Franklin Kuo’s “Protocols and Techniques for Data Communication Networks.”
“I read them and thought, ‘Wow, they can make computers talk to each other!’ I decided my career goal should be making computers communicate. It’s still what I do. I’m very much single minded!”
Armed with this well-defined ambition, Zhang received a master in electrical engineering, completing her degree with all A’s in 1981.
But Cal State had no PhD program in computer science, so Yano helped find schools where Zhang could seek a doctorate.
She applied to a dozen universities and was rejected by all her top choices, except one.
“Berkeley, Cal Tech, Stanford, they all said no, largely because I did so poorly on the English language section of the Graduate Records Exam,” she said. “But luckily MIT didn’t require GRE scores, and they admitted me.”
Early on in her stay at MIT, Zhang learned that one of the few other Chinese-born students was in a nearby building. She headed over to meet him, but along the way she met David Clark, then, as now, a key leader in Internet development.
“So,” Zhang said, “I bumped into Dave, and we talked. Eventually, he said to me, ‘Hey, look at this. It is hot off the presses. Read it and tell me what you think.’”
It was the TCP/IP specifications – fundamental protocols for Internet communication – which had just been published in September 1981.
“At first, I thought, what a strange thing. What is the meaning of this and why is it useful? Why do things work this way? I had so many questions,” Zhang said. “But now I’ve been reading it over 40 years, and I have so much appreciation for the design.
“And again, it’s the luck, pure luck that I met Clark, just as meeting Yano had been,” Zhang said.
With Clark’s support and guidance, Zhang was off to a fast start at MIT. Her very first published paper, “Why TCP Timers Don’t Work Well,” won the best student paper award from SIGCOMM under the Association for Computing Machinery – the first of many accolades she has received. And based on that paper, Zhang helped design a new transport protocol, NETBLT (Network Bulk Transfer).
Zhang spent eight years at MIT – long enough that, at one point, her mother feared she was becoming a professional student and urged her to graduate.
“Dave said, ‘What’s the rush? You can easily stay for two more years.’” Indeed, in those days, it was not uncommon to see some people stay in the PhD programs for 10 years.
And even with her PhD in hand, Zhang was encouraged by Clark to remain at the school as a postdoc.
But Zhang’s MIT officemate, Radia Perlman, herself very much an Internet pioneer, urged her to move on.
“She said, ‘Lixia, you’re crazy. You’ve been here eight years. Go, go as far as you can to see the rest of the world.’
“So I followed her suggestion and went across the country to the Xerox PARC.”
PARC wasn’t Zhang’s only job offer at the time; she would, for example, have earned a higher salary had she said yes to a job offer from Bell Labs.
But her mind was made up after an interview with PARC’s Computer Science Lab Manager Mark Weiser.
“I said, ‘Mark, why would you want me here? What do you want to get from me?’
“And he answered, ‘We don’t want anything from you. We want to help you grow into a first-class computer scientist.’”
And once again, Zhang said, it was heaven.
The year was 1989, barely a decade removed from rural labor in China, and Zhang now found herself at a well-funded, cutting edge R&D company that wanted her to become the best she could be – and to do so in style.
“It was during the golden years of PARC,” she said. “On my first day at work, the secretary, Lona, came in with design options – you know, those little squares and other shapes representing shelves, tables, chairs, cubicles. Lona said, ‘Lixia, how would you like to arrange your office? You can get anything you want.’
“And for my first trip at PARC, Lona ordered a limousine to take me to the airport! I couldn’t believe it.”
At PARC, Zhang continued her work on Internet architecture, shoulder-to-should with such other big names in networking as Van Jacobson, Steve Deering and Scott Shenker.
But in the mid 1990s the public world of computing began to change rapidly. The Internet was coming into its own.
“It went from being totally unknown to people on the street to the hottest topic of conversation. The worldwide web pushed the Internet into a new stage.”
Suddenly, universities were looking for people to teach networking, and there weren’t a lot of them. Zhang found herself with interview invitations from many universities, but one in particular got her attention.
Leonard Kleinrock, the UCLA professor famous for his pioneering work on the Arpanet, had asked David Clark to recommend a good name in networking, and Clark named Zhang.
“I was totally shocked when I picked up the phone and heard, ‘This is Len Kleinrock.’ So, I flew down to LA to interview, and that’s how I got to UCLA.”
Asked why she wanted to leave PARC for academics, Zhang cites another Chinese saying: “If you move a tree, the tree will die because it loses its roots. But if you move a person, the person will thrive.
“I’m a firm believer in that,” said Zhang who had spent six years at PARC. “When you move, you get a new environment, new people, new knowledge.”
Zhang also wanted to see how she might like teaching, which she had never done. And – no small thing – UCLA offered her instant tenure.
“So I went, and the plan was the same,” she said. “I thought I’d stay six or seven years, then move on – but that just didn’t happen! I think it’s because time accelerated; I didn’t realize how many years were passing.”
In her 25-plus years at UCLA, where she is the Jonathan B. Postel Professor of Computer Science, Zhang has continued to do important research. She is acclaimed for having created or contributed importantly to several communications protocols. She may be best known for inventing the term “Middle box” for the devices that sit between network elements, performing such functions as firewalls and load balancing to improve network security and performance.
But she has never neglected her role as a teacher.
“What I love most about my work is understanding, understanding how things work, what are the best ways to build systems. Teaching is not just passing your understanding to students, but more importantly to motivate them to ask more questions.”
Along the way, Zhang married a man she had met at Heilongjiang in China, a man who later moved to America and became a successful hardware engineer. Together, the couple has two sons, one a professor of computer science, the other finishing his MD residency.
In her free time, Zhang opts for the simple pleasure of reading books. Travel holds no special interest.
“With IETF meetings, NSF meetings, and project meetings, I traveled more than enough,” she said. “When I travel, I go from the airport to the conference hotel and back. So many meetings.
“East or West, home is best.”
Zhang says she has no personal ties in China – neither friends nor living relatives there. After her father died in China, her late mother came to live in Boston with Zhang’s brother, an investor.
For relaxation, Zhang swims, jogs – and she’s been a big NBA fan ever since her sons played basketball as youngsters.
“My whole family likes the Lakers,” she said. “We debate about who the real stars are.”
As for reading, Zhang’s tastes – no surprise – still gravitate to science.
Asked whether she ever reads fiction, Zhang said, “Fiction? Only in a gateway! I read bios. In particular I like Walter Isaacson’s books.”
Her literary hero is the late physicist and educator Richard Feynman.
“I read and reread everything by and about him: ‘Feynman Lectures on Physics,’ ‘Surely You’re Joking, Mr. Feynman,’ ‘What Do You Care What Other People Think?’ and many more. I probably have a full collection. My kids buy Feynman books as holiday gifts for me.”
Zhang said Feynman’s work not only taught her how to do research, but it is also profoundly relevant to the broader social challenges posed by the Internet.
For the past 13 years, she has been part of the multi-campus collaboration working on Named Data Networking (NDN), the radically different approach to network design originally proposed by Zhang’s long-time colleague, Van Jacobson.
Zhang believes NDN represents the most promising direction for mitigating the security challenges – hacks, fraud, fake identities, and now AI-powered deepfake – that plague today’s digital world. The problems, she notes, are made exponentially more urgent by Artificial Intelligence, and NDN aims to address them by replacing today’s Internet structure based on addresses with a new design utilizing semantically meaningful names that enable security to be built into the architecture, instead of being externally patched on.
“We still use a network design that was developed 40-plus years ago, when the so-called super computers had less capacity than an iPhone today. That was a great design at the time, but it just doesn’t fit today’s realities.”
“Think about when you were five and got a set of clothes. When you grew up, the clothes no longer fit. What we’re doing now is putting patches on the holes, but we need a new set of clothes. We need a new protocol architecture based on names.”
In support of this view, she cites a bit of apostolic wisdom, reprised by Feynman with regard to nuclear weapons:
“Feynman wrote: ‘To every man is given the key to the gates of heaven; the same key opens the gates of hell.’
“This applies to all technologies,” Zhang said. “Today’s Internet is no exception. In particular its lack of designed-in security makes it far too easy to be abused.”
But Zhang doesn’t spend much time worrying that the world will end up using its key the wrong way.
“As a researcher,” she said, “you don’t worry about things. You solve the problem. It’s our duty.
“So, I don’t worry. I’m working on solutions.”
