Robert Kahn

Coinventor of the transmission-control protocol and the Internet protocol

• Born: December 23, 1938
• Place of Birth: New York, New York

Primary Company/Organization: Bolt, Beranek, and Newman

Introduction

Robert Kahn's involvement with ARPANET and eventually the Internet began at the very start of the ARPANET program in 1968, when he was part of the team at Bolt, Beranek, and Newman that developed the network for the Department of Defense. Eventually joining the Advanced Research Projects Agency (ARPA) in 1973, Kahn headed what was the largest government computer research and development program to that time. Along with Vinton Cerf, Kahn developed the transmission-control protocol/Internet protocol (TCP/IP) in the early 1970s, establishing a means whereby computers on a network (such as ARPANET or the Internet) could communicate with computers on different networks. In addition, he was instrumental in creating open architecture, meaning that information could be accessed openly rather than through programs that were protected as proprietary intellectual property. Kahn's contributions have been recognized as among the most important to the creation of the Internet, earning him credit, along with Cerf, as one of the fathers of the Internet.

Early Life

Robert Kahn was born in New York City on December 23, 1938. He attended the City College of New York, where he earned a bachelor's degree in electrical engineering. Two years later, he received a master's degree from Princeton University, and two years after that he earned his doctorate from the same institution. Kahn's first job upon completing his education was at Bell Laboratories. He then became an assistant professor of electrical engineering at the Massachusetts Institute of Technology (MIT). In the late 1960s, he joined the staff at Bolt, Beranek, and Newman (BBN) in Cambridge, Massachusetts.

In late 1968, Larry Roberts, chief scientist and newly selected program manager at ARPA, released a request for proposal (RFP) in search of contractors to build a computer network for the agency, eventually to be known as ARPANET. BBN received the RFP and began to prepare a response in the form of a technical proposal. Kahn was assigned to the team that prepared this response.

In early 1969, BBN was announced as the winner and thus began to develop the network that would eventually result in computers (referred to as interface message processors, or IMPs) at four different sites: the University of California at Los Angeles, the University of California at Santa Barbara, Stanford University, and the University of Utah.

Kahn's involvement, which began with developing the proposal, would continue into 1971. He not only would help in the development but also had an influence on the approach that would be taken to design the architecture of the network. His view, which would be incorporated into the overall development, was that however the network was to be configured, it ought to be planned in such a way that its basic organization would remain the same even as the network expanded. In that way, costly changes in the future could be avoided.

As time went on, Kahn's involvement in the project was accompanied by a growing conviction that the network was increasingly important, a feeling that he did not believe was being shared by BBN, which tended to develop solutions and then move on to the next problem. In 1972, Kahn was offered a position as head of ARPA's Information Processing Techniques Office (IPTO). Later that year, Kahn gave a demonstration at an international computing conference of ARPANET's capabilities, linking forty computers simultaneously.

Life's Work

While Kahn's work up to 1972 was important in defining and developing networks, his major contributions would be made in the later 1970s. First, as IPTO director, he started and managed the largest computer development program to that time, the Strategic Computing Program. Second, he and Vinton Cerf developed a solution to the problem of how networks might communicate with one another despite differences in equipment and architecture.

The computers connected to ARPANET had all been brought in as integral parts of the ARPA network. There was no problem with their ability to communicate among themselves. What would happen, however, if there was a need (and there certainly would be at some point) for ARPANET to communicate with computers on other networks? For the time being, there was no commonality in hardware or software, and consequently none in the ability of networks to contact and understand one another. That was the problem that Kahn and Cerf would solve.

Their eventual solution was to develop network protocols. A protocol is software or hardware that defines how communications are conducted so that dissimilar networks can communicate with each other. Protocols handle communications and the rules at different levels. One protocol will handle how different hardware items will interact. Another will establish how messages will be passed back and forth (the packet-switching method that Leonard Kleinrock claimed to have developed was the means of conveying messages on ARPANET and later the Internet). Another may govern how browsers will receive messages, while yet another may resolve lost packet issues or authentication.

In the spring of 1973, Cerf joined Kahn on the project. They started by conducting research on reliable data communications across packet radio networks, factored in lessons learned from the networking control protocol, and then created the next-generation transmission-control protocol (TCP), the standard protocol used on the Internet today. In the early versions of this technology, there was only one core protocol, which was named TCP. At the time, these letters did not stand for what they do today, transmission-control protocol, but instead stood for the transmission-control program. The first version of this predecessor of modern TCP was written in 1973, then revised and formally documented in RFC 675, “Specification of Internet Transmission Control Program” (December 1974; RFC is a standard designation in computer and Internet engineering meaning “request for comments”).

What Kahn and Cerf developed, then, was to be known as the transmission-control protocol/Internet protocol (TCP/IP). These two layers form the middle of a four-layer structure that makes communication on the Internet possible. The first is known as a link layer and provides the means and the rules for communications within a specific network. The next is the Internet protocol (IP) layer, developed by Cerf and Kahn, which makes it possible for different local networks to connect and communicate. Once a connection is established, the transport layer (Kahn and Cerf's TCP) allows the host in each network to connect and communicate with the host in another network. Finally, there is an application layer, which governs how data service processes are run (for example, a web browser interacting on a machine with a web server located elsewhere).

What Kahn and Cerf did was to make it possible for individuals on different networks to communicate with one another. While this was extremely important in resolving an immediate need, its importance went beyond a technical solution. By insisting that the architecture and the protocols be open to everyone, they ensured that these protocols could be made to work anywhere by anyone and not subjected to the restrictions that would occur when using proprietary systems. The Department of Defense, even at this early stage, was making decisions regarding programs such as the Worldwide Military Command and Control System (in which proprietary solutions would hinder open communications). Kahn and Cerf's solution ensured that similar problems would not exist within ARPANET and subsequently the Internet.

Today's IP networking represents a synthesis of several developments that began to evolve in the 1960s and 1970s, namely the Internet and local area networks, or LANs (which emerged in the mid- to late-1980s), together with the advent of the World Wide Web in the early 1990s.

Personal Life

Kahn left ARPA in 1986, thereafter serving as chief executive officer (CEO) and president of the Corporation for National Research Initiatives (CNRI), a nonprofit organization that seeks to find solutions to developing a national Information infrastructure. He has received many awards, including the Association for Computing Machinery's Alan M. Turing Award and its President's Award. In conjunction with his collaborator, Cerf, Kahn received the US National Medal of Technology, the Marconi Award, and the Presidential Medal of Freedom. He shared the Charles Stark Draper Prize for 2001 with Cerf, Leonard Kleinrock, and Larry Roberts for their work on the ARPANET and Internet. In addition, Kahn has received many honorary degrees.

Kahn's cousin, Herman Kahn (1922–83), was a military analyst and theorist who made a specialty of studying the possibilities of atomic warfare and its results, publishing two books on the subject, On Thermonuclear War and On Escalation. Herman Kahn is said by some to have been one of the models for the titular character in the film Dr. Strangelove.

Bibliography

Beranek, Leo. “Roots of the Internet: A Personal History.” Massachusetts Historical Review 2 (2000): 55–75. Print.

Denning, Peter J., and Robert E. Kahn. “The Long Quest for Universal Information Access.” Communications of the ACM 53.12 (2010): 34–36. Business Source Complete. Web. 22 July 2012.

Hafner, Katie. Where Wizards Stay Up Late: The Origins of the Internet. New York: Simon, 1996. Print.

Salus, Peter H. Casting the Net: From ARPANET to Internet and Beyond. Reading: Addison, 1995. Print.