Unsung innovators: Robert Kahn, the ‘stepfather’ of the Internet – Computerworld

Americas

  • United States

Asia

Unsung innovators: Robert Kahn, the ‘stepfather’ of the Internet

feature
Dec 03, 20076 mins
E-commerce SoftwareGovernment ITIT Leadership

Bob Kahn won’t take the bait when asked if he is the father of the Internet. He acts almost as if it’s a trick question, and it isn’t the first time he’s heard it.

“If I said I did invent it, I’d likely end up in a cartoon somewhere,” says Kahn. Still, he is the man responsible for the initial system design of the Arpanet, and he helped convince Vint Cerf, the one who usually gets “father of the Internet” status, to come to work with him at the Defense Advanced Research Projects Agency (DARPA).

Initially, in 1973, DARPA contracted with Cerf to work with Kahn on aspects of the nascent Internet project, “after I had the basic idea down,” to help complete the design and implementation of the TCP (Transmission Control Protocol), Kahn says. The idea was to replace the Arpanet’s by-then-outdated method of passing along messages, called the Network Control Protocol.

Together, Kahn and Cerf published their now-famous paper, “A Protocol for Packet Network Intercommunication.” The two created what became the lingua franca of the Net — TCP/IP — which has been used to transmit information over the Internet ever since.

The Kahn and Arpanet story goes back to October 1966, when Kahn was taking a break from his professorship at MIT and working at Bolt Beranek and Newman (BBN). Back then, BBN was primarily an architectural acoustics research firm in Cambridge, Mass.

While there, Kahn took the lead in writing a technical proposal to submit to DARPA describing how to implement a packet-switching network, soon to be known as the Arpanet. The proposal later won the bid from DARPA (then called ARPA).

“BBN ended up getting the contract to build the Arpanet, and a four-node version got deployed and [was] working in December 1969,” Kahn explains. Those first nodes linked computers at UCLA, the Stanford Research Institute, the University of California at Santa Barbara and the University of Utah. But it took almost three more years for the research community to get those host computers connected to the Net. The first public demonstration of the Arpanet was in 1972, when approximately 50 nodes were operating, he says.

Robert Kahn, 'stepfather' of the Internet

Robert Kahn, ‘step-father’ of the InternetGiven his connection to the networking world, Kahn thought it was unfortunate that the press misquoted Vice President Al Gore in the 1990s as saying he invented the Internet.

“I remember getting about 200 calls from the press that week, all asking roughly the same question: ‘Are you angry?’ No, I certainly wasn’t angry or anything close to that frame of mind.”

Gore did a lot of good for the Internet that deserves recognition, Kahn says, including promoting the Internet in schools and to citizens in a very public way. But he was treated very poorly for what was at worst an inappropriate choice of words, Kahn explains.

Cerf and Kahn both became widely known in technical circles as the DARPA guys behind the Internet. Cerf got much of the mainstream press credit, though. Kahn says he isn’t bitter and that it’s easy to understand. “It’s funny; Vint lights up when he’s in front of the camera,” says Kahn with a laugh. “He’s great with the press.”

That, and a major public relations push by both MCI and later Google — Cerf’s former and current employer, respectively — perhaps explains why Cerf is the more well-known name, at least at this point in time.

As Kahn points out, that could still change.

The two will go down in history as an inseparable team. In 2005, both won the coveted Turing Award, and both were awarded the National Medal of Technology by President Clinton in 1997 and the Presidential Medal of Freedom by President Bush in 2005.

Kahn finds it funny that, looking back, people think getting the funding for a program to do the Internet and making it happen was “a slam-dunk.”

“I’ll tell you, the only reason we succeeded in building it was that no one cared enough in the early 1970s to get involved and try to stop it,” he says. “So few companies had time-sharing computers back then, and thus, there wasn’t seen to be much of a market for interactive computer networks. And so there was no effort in the private sector to interconnect nonexistent networks to each other.”

DARPA happened to have three such networks in existence or in development, Kahn says, “so we could imagine the problem and attempt to solve the interconnection problem. It was not until later that alternate networks, such as the Ethernet and personal computers, emerged in the commercial marketplace. That made a big difference.”

One big moment that isn’t often recognized, he says, is when DARPA — working with a number of contractors, including Collins Radio, BBN and others — demonstrated the first successful TCP connection traversing three dissimilar but interconnected networks. November 22, 2007, marked the 30th anniversary of that demo.

It was an event that Kahn believes should be more recognized because, like today’s Internet, it was the “internetworking” of three disparate networks, and it foreshadowed the cellular and Wi-Fi systems of today.

Kahn and Cerf also teamed up to run the then-fledgling Internet Engineering Task Force, which originally was created to manage short-term items to “get the Internet up and running,” and later took on responsibility for other Internet standards groups.

After his stint at DARPA, Kahn didn’t stop pioneering. In 1986, he started the Corporation for National Research Initiatives (CNRI). The Reston, Va.-based organization helps shepherd various technology infrastructure projects. With funding from the National Science Foundation and DARPA, the CNRI helped create the first Gigabit networks operating at speeds above 1 billion bits per second. The CNRI also funded the development of Mosaic, the first popular Web browser, at the University of Illinois.

Among CNRI’s other achievements is the technology around the digital object architecture, which, Kahn explains, “was an effort on my part to reinvent the Internet around the idea of managing digital objects instead of just moving bits around.”

Of course, scores of people at DARPA and elsewhere also deserve credit for developing the Arpanet, which was the first packet-switched network and was one of the three original networks linked via the Internet, along with a satellite network and a radio network. But Kahn has been there from the very beginning and is still involved today.