Vint Cerf

By: Eli Regalado Issue: Big Ideas, Smart People Section: Business Vint CerfIt doesn’t matter if you’re bored, researching, or downloading music—if you’ve ever "surfed" the web, you can thank Vinton Cerf. Vint, as he is called, is one of the forefathers of the Internet and is tied to engineering the original saleable email. This Stanford graduate continues to thrive within the world of the Internet as an employee of Google. He has seen the beginning, he has seen what it has become, and he will continue to design where it will go. With almost two billion people connected to the Internet worldwide, there is plenty of room for improvement and Vint Cerf is helping lead the charge.

ICOSA: As the Chief Internet Evangelist at Google, what are you doing now?

CERF: Since October 2005, I have been Google’s Chief Internet Evangelist which means that my job is to stay on top of the technological evolution of the Internet and the applications that are rapidly evolving on it. I also deal with domestic and international policy issues. There is still a continued effort to get more Internet built because only about 25-30 percent of the population is online at this point. I’d like everyone—all 6.6 billion of us—to be up on the Net.

I’ve observed that mobiles have become a very significant part of the Internet’s environment. Today, there are about 4.5 billion mobiles in use, but not all of them are Internet enabled. However, a reasonable percentage of them —maybe 20 percent—are. As time goes on, more and more phones will have the ability to make use of the Internet and the World Wide Web through mobile applications. So, I see a kind of rapid advance on the mobile side to bring people up on the net and a slower, or at least delayed, introduction of broadband services.

ICOSA: In its earliest stages the Internet was an experiment—when you started where did you see it going and did you have any idea it was going to be as big as it is today?

CERF: The simple answer is no. We had no idea the Internet would be what it is today. ...Well that’s not entirely fair and I need to give you a little context in order to appreciate that. In 1973, Bob Kahn and I started designing what we call the Internet today. It was built on the strengths of an experiment that was done four years before, called the Arpanet (Arpa), which was a network based on dedicated telephone circuits linking specialized packet switching computers together. As graduate students and engineers, we were developing this technology as we were using it; we were making use of remote time share access to machines in the same kind of way that a web server is used today remotely on the net. Keep in mind the World Wide Web hadn’t been invented yet, so it was a much simpler interface, but the whole idea and motivation was resource sharing, the ability to share other people’s software and share people’s data on machines that were located elsewhere.

More importantly, around the late 1960’s a man by the name of Douglas Englebart had undertaken an interesting project at Stanford Research Institute International in Menlo Park, California called the augmentation of human intellect. The idea was that computers would be able to augment our human intelligence by performing repetitive functions, indexing or doing other kinds of tasks with large amounts of data that we couldn’t otherwise do ourselves. Englebart’s group was interested in information sharing, document creation and sharing, and from that he invented the notion of hyperlinking. There was a lot of collaborative technology. He invented a portrait mode presentation that looked like a black and white piece of paper as opposed to the yellow and green screens of the day. It was an extremely advanced piece of work that was all taking place in the late 1960’s.

Also around 1970, Xerox started a research organization in Palo Alto called Palo Alto Research Center (PARC), which brought some of the smartest people together—many from the ARPA sponsored programs. Back in ’73 and ’74, everyone at PARC had their own personal computer. These machines were $50,000 each. It was at this time, Bob Metcalf, a member of PARC went to an ARPA project at the University of Hawaii (ALOHAnet). ALOHAnet was a radio-based network where terminals on the outer islands were transmitting data back to Oahu where the University of Hawaii is located using a very dynamic spectrum sharing technique. Well, Metcalf went back to PARC and said, "I can do this on a coaxial cable and I’m going to call it Ethernet." So, he’s inventing Ethernet in 1973, linking all these Alto machines together at high speed. At the same time, I was a mile and a half away at Stanford University inventing TCP/IP with Bob Kahn. Electronic mail had also been invented for the Arpanet during the same time period. So, we’re all using email; Metcalf is doing the Ethernet and the Alto personal computers; I was doing TCP/IP Internet architecture with Bob Kahn—and by the launch of this thing in early 1983—we all have had over a decade of experience in computer networking in the form of the Arpanet.

At the same time, personal computers were starting to emerge in the private sector and the Ethernet had become publically available as a product, proving there was a growing appreciation for the capability of the system. At a point when we rolled out the Internet in 1983, we knew we could do packetized voice and video on satellites, on the radio, and it could work on the Arpanet and Ethernet.

Now, none of this was commercial and it didn’t become commercial until around 1989. I realized in 1988, however, that the only way it would ever become available to the general public was if it could be made to be commercially self-supported. So, I went on a campaign to get the rights to carry commercial traffic on the government sponsored backbones of the Internet. I got that permission in late 1988, and in 1989, some of my colleagues and I put up a gateway linking the Internet email system, which had been around since 1971, with the MCI Mail System that had been built and rolled out in 1983. Just five years later we started to link this commercial email system with the academic Internet.

In 1989, three commercial Internet services came along and about that same time, Tim Berners Lee invented the World Wide Web, which started to show up in the form of the Mosaic Browser in 1992. By the time the mid-1990’s rolled around, we were in the dot-com boom period. The point is that during the course of nearly 20 years of evolution—from the early 1970’s to mid-1990’s—we saw repeated opportunities for the Internet to increase its utility, until finally it became available to the public.

ICOSA: While the Internet was being developed, did anyone think you were nuts for working on this project?

CERF: Well, certainly in the earliest period during the Arpanet phase, the conventional wisdom for telecom was that the circuit switching technology used in the telephone system was the only way to do this type of communication. We were soundly criticized for attempting to do this "silly" thing by AT&T and others who had been in the telecom game for a while. We said, "Well, we are going to do it anyway." And they said, "We’re happy to lease you the circuits and we’ll let you do this ‘silly’ thing." Of course later on it actually worked and they didn’t’ think it was so "silly" then. This "silly" thing was a paradigm breaker because packet switching was dramatically different from circuit switching technology. Now the technologies support the exploration of the solar system with manned and robotic space craft and so many other things.

ICOSA: In your opinion, how does rapid information sharing affect modern society?

CERF: There are positives and negative aspects of rapid information sharing. On the good side, we learn about events much more quickly and in ways we wouldn’t have before because they are being indexed by our search engines. On the negative side, bad things happen around the world and when we become aware of them, it feels like it’s local. So, when negative things happen you get this uneasy feeling that bad things are happening everywhere, including where you are, even if that is not true. The immediacy of the information gives that feeling.

While television networks like CNN and FOX have those same characteristics, the Internet exacerbates it. People tweeting, taking video and posting to YouTube, and sending emails increase the amount of interaction and awareness of what is going on in the world. We have seen potential political ramifications of this especially in the post-Iran election period. Videos were posted to YouTube and tweets were picked up on the user’s mobile that painted a picture for the rest of the world about what was going on there—we’re seeing this rapidly growing network giving voice to things that otherwise would have been invisible.

Also on the negative side, the network is a bearer of terror. It is used by groups that have inimical agendas; whether it is terrorists, people committing fraud, or other abuses on the Net, it becomes a conveyor to those anti-social activities. The benefits seem to outweigh the deficits, yet we are hard pressed as countries to find ways of protecting access to the network while also doing something about people who are abusing it.

ICOSA: What is your view on Internet television and where do you think it is going?

CERF: We are seeing a fairly quick convergence of all the various means of delivering video coming together—in some cases they are converging on set top boxes and in some cases coming out of multiple boxes. For example, DVD and Blue-Ray players have recently become Internet capable, so you can go to Netflix or YouTube through those players and control your experience with a remote. Or, you might actually use your laptop to surf the Internet, find videos that can be streamed, and then output it all onto a high resolution large screen display.

It’s been a very interesting phenomenon watching separate networking and media come together in one presentation device. What is happening now is that all three of these different kinds of communication networks—voice, television, and Internet—are becoming commonly interconnected, and all the different media that can be carried on them is becoming interconnected. What used to take weeks now takes minutes or seconds.

ICOSA: What are you working on now that is changing the way things are done?

CERF: Back in 1998, I started working on the design of an interplanetary extension of the Internet—something that would allow us to operate networks across the solar system. The first reaction was, "This is just silly. It’s science fiction. And, I hope you are o.k. in your little padded room." However, just over 10 years later this is serious engineering. We have already done testing in deep space with the new protocols that will work over huge distances. The purpose of the technology is to support the exploration of the solar system with manned and robotic space craft. It will also provide a richer networking environment so that information can be obtained from the robotic devices—like the rovers and the orbiters—that are exploring the surfaces of planets or are looking down on activities of asteroids or comets. And, of course, this technology will help astronauts interact not only with each other, but with Houston back on earth. Because the networking capabilities up until now for space communications have been point-to-point radio links, my colleagues and I have been working to make the network as rich as the Internet, in terms of being able to easily allow multiple devices to be part of the interplanetary communication network.

And, we at Google are heavily invested in cloud computing systems. My academic colleagues are irked when I say, "Gee it’s time-sharing all over again." They argue, "Well it’s more than that." So, to be fair, that’s probably right—in a sense.

The computing utility used to be displayed as this giant building somewhere in the middle of the country with steam coming out of the top and a bunch of telephone lines connected to it. Today, it’s a bunch of big buildings with a bunch of steam coming out of the top with essentially laptops all stacked up on top of each other and a whole bunch of Internet connections going into it. We’ve reinvented big computing utility. It’s dramatically different in its character. There are a lot of very useful notions of cloud computing to do business, and it also presents the most interesting problem—getting the clouds to talk to each other. Here we are in 2011, and inter-cloud interactions are in a state where we were with the Internet in 1973. For somebody, this inter-cloud stuff is going to be another incredible experience.