A crystal radio in Manhattan
Leonard Kleinrock was born in New York City on June 13, 1934, to a Jewish family. When he was six, he found the plans for a crystal radio printed in the centre pages of a Superman comic book. He collected the parts himself: a used razor blade of his father’s, pencil lead, a cardboard tube, some wire, and an earphone taken from a public telephone. He still needed a variable capacitor, so his mother took him to the electronics shops of Canal Street, where he had to admit to the clerk that he did not know what size he needed or what it was for. When the finished radio played music with no batteries and no power supply, he later wrote, he became an engineer at heart.
He went to the Bronx High School of Science, graduating in 1951, and took radio engineering courses on the side.
Kleinrock has often told the radio story as the beginning of everything that followed: a child discovering that invisible signals could be pulled out of the air with a few scraps and some curiosity. The same fascination with how information travels, and how it can be made to travel better, ran through his whole career.
Night school and MIT
Even the tuition-free City College of New York was more than his family could afford if he studied full time, so Kleinrock enrolled in its evening session while working full time as an electronics technician to help support his family. It took him about five and a half years. He graduated in 1957 with a bachelor’s degree in electrical engineering, first in his class, and had been elected president of the evening session’s student body.
His record won him a full fellowship to the Massachusetts Institute of Technology. Most of his classmates were drawn to the fashionable field of information theory. Kleinrock chose something almost no one was studying: how data could flow through networks of communicating computers. He earned his master’s degree in 1959 and his doctorate in 1963.
It was an unusual choice. Data networks barely existed, and computers of the day were enormous, expensive machines that rarely communicated with one another. Kleinrock reasoned that as computers multiplied they would need to share resources and exchange information, and that someone would have to work out how such networks should be designed.
The mathematics of waiting in line
In May 1961 Kleinrock submitted a doctoral proposal titled Information Flow in Large Communication Nets. His thesis, completed in 1962 and published in 1964 as the book Communication Nets, applied queueing theory, the mathematics of waiting lines, to networks in which messages are stored at each node and forwarded when a line is free. He studied how delay depends on the capacity of links, the rules for priority, the choice of routes and the shape of the network, and he analysed a real message-switching system that Western Union used for telegrams.
The work gave engineers tools to predict how a network would perform before it was built. When the British computer scientist Donald Davies, one of the inventors of packet switching, wrote his 1966 proposal for a packet-switched network, he used Kleinrock’s techniques to show that such a network could respond fast enough for human users. Kleinrock’s two-volume Queueing Systems, published in 1975 and 1976, became standard texts.
The first node of the ARPANET
Kleinrock joined the faculty of the University of California, Los Angeles, in 1963, where he has remained ever since. In 1967 his MIT contemporary Larry Roberts, who was planning a computer network for the U.S. Defense Department’s Advanced Research Projects Agency, brought him into the project, and Kleinrock helped write the functional specification for the ARPANET in 1968. That year he won a contract to establish the Network Measurement Center at UCLA, to measure and model the performance of the new network, and UCLA was chosen as its first node.
The first Interface Message Processor, the packet switch built by the firm Bolt, Beranek and Newman, arrived at UCLA on August 30, 1969, and on September 2 bits flowed between it and UCLA’s host computer, the day Kleinrock calls the infant internet’s first breath of life. His team, which included the graduate students Steve Crocker, Jon Postel and Vint Cerf, wrote the host software and protocols that let computers talk across the network.
“Lo”
At 10:30 in the evening of October 29, 1969, in room 3420 of Boelter Hall at UCLA, a student programmer, Charley Kline, working under Kleinrock’s supervision, tried to log in to a computer at the Stanford Research Institute, several hundred miles to the north. He typed an L and an O, each confirmed by telephone. When he typed G, the system at Stanford crashed. The first message ever sent over the network that became the internet was therefore “lo”. About an hour later the connection was restored and the full login succeeded.
By December 1969 four sites were connected, and by mid-1970 ten nodes spanned the United States. Kleinrock’s measurement centre deliberately stressed the young network to find its weaknesses, uncovering failures such as deadlocks that his team nicknamed the Christmas lockup and the piggyback lockup, and it helped fix them. Email soon became the network’s most popular use.
From ARPANET to the internet
Kleinrock was also among the first to use the network for everyday purposes. In 1973, while Larry Roberts was in England, Kleinrock used the ARPANET to exchange messages with him in real time, an early example of the kind of instant communication that later became universal. In the mid-1970s UCLA used the network to control a satellite in orbit over the Atlantic, a demonstration of how far the idea of remote computing could reach.
Through the 1970s Kleinrock and his students analysed packet-switched networks, packet radio, satellite links and local area networks. His work with his student Farouk Kamoun on hierarchical routing, published in 1977, showed how very large networks could keep routing manageable by grouping nodes into levels, an idea that remains central to how the internet operates. Many of his doctoral students, about fifty in all, went on to shape networking in industry and academia.
He also helped make the case for a national network. In 1988 he chaired a National Research Council committee whose report, Toward a National Research Network, told Congress there was a clear and urgent need for one, and he testified before Senator Al Gore. The report influenced the High Performance Computing Act of 1991, whose funding helped support the development of the Mosaic web browser, which in turn sped the spread of the World Wide Web. In 1994 he led the committee that produced Realizing the Information Future: The Internet and Beyond.
Credit and controversy
Kleinrock has described his early work as establishing a mathematical theory of packet networks and has sometimes been called a father of the internet. That claim has been disputed. Other pioneers, including Paul Baran, Donald Davies and Robert Taylor, argued that his doctoral work modelled message switching rather than inventing packet switching, and historians and the National Inventors Hall of Fame credit Baran and Davies with independently inventing packet switching itself. What is not in dispute is that Kleinrock’s queueing models became the standard way to analyse network performance, and that his laboratory was where the ARPANET came to life.
He has also been an entrepreneur: he was a co-founder of Linkabit, the communications company started with Irwin Jacobs and Andrew Viterbi, and he founded the network company Nomadix and the Technology Transfer Institute.
Honours and Jewish roots
Kleinrock was elected to the National Academy of Engineering in 1980, one of its youngest members at the time. He shared the Charles Stark Draper Prize in 2001 with Vint Cerf, Robert Kahn and Larry Roberts for the development of the internet, received the National Medal of Science for 2007 from President George W. Bush at the White House in 2008, shared the Dan David Prize in 2010, and was inducted into the Internet Hall of Fame in 2012. In 2011 UCLA restored room 3420 of Boelter Hall to its 1969 condition as the Kleinrock Internet Heritage Site and Archive.
Kleinrock was born into a Jewish family in New York and came up through the city’s free public institutions, Bronx Science and City College, which carried so many children of Jewish families into science and engineering. His own accounts of his life focus on his family’s modest means and his education rather than on religious practice.
Legacy
Leonard Kleinrock helped turn the idea of a computer network from a sketch into a working system that could be measured, understood and improved. His mathematics told engineers how networks would behave; his laboratory hosted the first node and sent the first message; and his students wrote software and went on to build the protocols and companies of the networked world.
In recognition of that role, the restored room in Boelter Hall, with an original Interface Message Processor on display, now welcomes visitors as a place where the history of the internet began.
Today more than five billion people use the internet to learn, work, trade and stay in touch. The two letters typed in a UCLA lab on an October night in 1969 were the first words of that conversation.
Sources & further reading
Leonard Kleinrock’s Personal History/Biography: The Birth of the Internet, UCLA · Leonard Kleinrock, Internet Hall of Fame · Leonard Kleinrock, Wikipedia · Jewish Computer and Information Scientists, jinfo.org · Researched from the sources above and checked against them. Spotted an error? Tell us.
A life of tikkun olam. This is part of tikkun olam, the Jewish idea of repairing the world, and every one of us can add to it.