He is this era's Alexander Graham Bell. In the early 1970s, Vint Cerf, along with frequent collaborator Bob Kahn, wrote a paper sketching out a way that different computer networks might speak to one another. By 1975, a group working under Cerf at his alma mater, Stanford, had developed a language to enable the inter-network communication imagined in the paper—and, lo, the Transmission Control Protocol was born. The Internet couldn't exist without its lingua franca, the so-called TCP/IP.
It wasn't immediately popular. Not until 1985—exactly 30 years ago this month—did a Boston tech company become the first to register a domain name on this new network of networks. Only five other organizations sought a home there that entire first year. But, of course, soon enough it would experience explosive growth.
Now 71 years old, Cerf still influences tech's direction. He is Google's Chief Internet Evangelist, and he consults with politicians and policymakers around the world on laws affecting the Net. But little in his storied career can compare to the first time the TCP/IP language helped computers attached to different networks to communicate across vast distances—it was something of a Bell moment.
"There was a van in the San Francisco Bay Area that sent, by radio, an electronic packet to the Arpanet," he recalls, speaking of a primitive wired network developed by scientists with U.S. military funds. "It went by satellite to Norway, then to University College, London, then back by satellite over the Atlantic, through the Arpanet's wires again, and then to computers at the University of Southern California. I mean, it just went from San Francisco to Los Angeles, but it crossed 100,000 miles to do so."
There's still some wonder in his voice. Interviewed in a relatively generic office in Google's Virginia campus, his white beard well trimmed, Cerf is nattily attired in a blue three-piece suit, a formality rare in slovenly tech circles. He explains that he wasn't always so proper—he operated in the thick of the grubby, anarchic, forever-in-blue-jeans community that came up with many of the key aspects of what would become the Internet in the '70s and '80s. "We depended on input, on a real sense of collaboration," he says. "It was a bit…chaotic."
Then he moved to Washington, D.C. "When I left Stanford to join DARPA"—the Arpanet's administrator—"my wife said I should wear a suit and found some for me at Saks Fifth Avenue. When I testified before Congress, the committee chair sent a note to our director thanking me for my testimony and saying, by the way, I was the best-dressed representative of DARPA they'd ever seen."
This adopted smoothness proved helpful when a competitor to TCP/IP, the so-called Open Systems Interconnection, emerged. OSI was popular in Europe—produced, the story goes, in a more orderly fashion by scientists working at the behest of bureaucrats. To many, the protocol felt more secure.
As DARPA's representative, Cerf had the unpleasant duty of attending OSI conferences, biting his tongue, cordially disagreeing that the upstart was all that. "I was the guy forever writing the counter-paper," he says. In those times, the soft-spoken Cerf persuaded many key people that TCP/IP had more going for it—more flexibility, fewer bugs—and the vogue for OSI soon faded.
For a top scientist, Cerf is adept at translating the complex into terms that even the technologically illiterate can assimilate. He makes the present seem comprehensible, the future not so scary.
"Some fairly notable people have worried that artificial intelligence is becoming a hazard to humanity," he says, addressing reported comments by the likes of scientist Stephen Hawking and Tesla founder Elon Musk. "But most AI is being used as a co-operative measure, not a coercive one. Isaac Asimov's laws of robotics haven't been codified, it is true, but they are generally being observed." (Among other things, the laws articulated in the early science-fiction author's work ban robots from harming humans.)
One example of human-machine co-operation Cerf cites is the Google Glass—this, although his company recently decided to take its controversial wearable computer off the market. "The Glass brings computers into ordinary human discourse, with machines hearing and seeing what you are hearing and seeing, and becoming more of a party…participating in the conversation."
If he remains gee-whiz about tech's possibilities—"Imagine, the phone can take an image of a menu written in another language and translate it into yours"—he also grasps common worries. "I'm interested in the Internet of Things, and we have lots of smart-home technology—a Nest Thermostat, a sensor that gathers information about heating, humidity and lighting in each room of our house. That said, the data collected about who's at home, where they are…could be abused by bad guys."
Talk of bad guys leads him to parse the Sony hack. "The issue is not only the information that was released, but the corporate information that was damaged and destroyed." A paradox: Google tries to make its software resistant to such invasions by releasing its coding to the public. "There's a bounty for those who find bugs or potential hazards." This openness, a collaboration with the many, not the few—these are holdovers from the Internet's early days.
What, precisely, does Google ask its "Internet Evangelist" to do? "I push for more Internet—the expansion of broadband, increased access in the developing world, maybe through drones, high-altitude balloons." Cerf has advocated hard for net neutrality, pushing (again before a congressional committee) to stop well-to-do corporations, like his employer, from getting better, faster service than small fry do.
Among Cerf's many awards is one named for Alan Turing, whose early computer helped break the Nazi Enigma code, thereby shortening the Second World War. A fictional version of Turing is, of course, the central character in The Imitation Game, one of this year's Oscar nominees. Cerf is offended at the liberties the filmmakers took with the science. Of course he is. "I felt that the movie under-explained the significance of Turing's work on computing. His post-World War II contributions were technically as important and perhaps more pervasive than cracking [the code]."
Cerf once said of TCP/IP's early success, "To borrow a phrase, now the Internet could go where no network had gone before." Did he ever think it would go this big, this quickly—this, well, boldly? "No. So much about it has been surprising. The explosive adoption of the World Wide Web protocols. The growth of hacking, phishing and other hazards. The hypergolic confluence of the smartphone and the Internet. Some of it, we imagined, yes, but much of it—no."