The Grill: Google's Internet Evangelist Vint Cerf on the Hot Seat

Google's chief Internet evangelist talks about the sorry state of online security, his hopes for deep space communication and using IT to find saffron in the wilderness

Vint Cerf is the co-designer with Robert Kahn of the TCP/IP protocols and the basic architecture of the Internet. In 2005, he and Kahn received the highest civilian honor bestowed in the U.S., the Presidential Medal of Freedom.

As one of the founding fathers of the Internet, what about your creation makes you most proud? It's the commercialization of the network that has given it its greatest character. Otherwise, it would be academic and military.

The commercial incarnation didn't really happen until 1989, which makes the Internet 18 years old. It's had an awkward [adolescence]. It still isn't quite clear how it's going to fit into the world. It's boundaryless in many respects, so both good things and bad things manifest. I'm very impressed by the quantity and quality of content on the Net, [but] there's some pretty bad content as well.

Dossier

Steve Mills
Name: Vinton G. CerfTitle: Vice president and chief Internet evangelistCompany: Google Inc.Location: Herndon, Va.Favorite nonwork pastime: Reading Most interesting thing people don't know about him: "I used to play the cello. I was invited to attend a master class led by Pau Casals. I was 15."Philosophy in a nutshell: "You can learn something from everyone; optimism is better than pessimism; most people rise to your level of expectation."Favorite vice: Häagen-Dazs coffee ice cream. Dream dinner table guests: "Bill and Hillary Clinton; Eric Lander of the Whitehead Institute and his wife; geneticist Richard Dawkins, one of the most outspoken Darwinists in our contemporary period; and of course, Sigrid [Cerf's wife]."

How do you use the Internet? My personal use of the Internet is heavily oriented toward finding information, and I'm just astonished by the information available.

I was on vacation driving around in Utah and Arizona in remote places. We were planning to prepare some paella and needed saffron. We were out in the middle of nowhere, but I was picking up a good, strong signal on my BlackBerry. I went online and got a list of stores. We called using the mobile [phone]. Five minutes after going online and finding a store and phoning, I bought $13 worth of saffron. I thought, "By God! I can't believe what we just did!"

How has the Internet affected mobility and vice versa? If you know you'll have access wherever you go, you change some of your habits. I'm an avid laptop user, and I just spent two weeks without my laptop. I've certainly not been without a computer for that [length of] time [before]. I had only my BlackBerry, and I survived.

I'm beginning to think that some of the stories in Star Trek about these giant galactic databases containing millions of years of knowledge were sort of on the dim threshold of how to take advantage of information in machine-processable form. But there's a big technical issue that has me worried.

What's that? The information on the Net is not all simple text. It's structured, whether it's Microsoft Word documents or PDFs. That means the information is only really accessible if you understand how to interpret the bits. What happens when files are there and we don't know how to interpret them anymore?

How would we lose that ability? If you have a CD but the form isn't known anymore. I have 5 1/4-in. diskettes, but nothing to read them. Even 3 1/2-in. diskette readers are becoming hard to come by. The physical source media change. We may lose the ability to read them. People forget what they stored on the Net. We can't necessarily guarantee people will upgrade instances of things.

What's your assessment of security on the Internet? Internet security is really not in great shape right now. There are many avenues of attack available to people who want to cause trouble. DNS is the most serious kind, and the bots that carry out those denials of service. It's too easy to compromise laptops because operating system software is vulnerable, and this is true for just about every operating system. The problem isn't just operating system vulnerabilities. If you download a piece of Java code, it may be able to get access to all sorts of critical data on a secure computer. It's that kind of risk and vulnerability that worries me a great deal. If we can't do a great job, people will be less comfortable using the Net and computers.

You've been married to your wife, Sigrid, for more than 40 years. You also have a rich intellectual and professional life. What's your secret to happiness? You'd find my wife, Sigrid, to be extraordinary. It is she who has been very accommodating of my workaholism over the last 40 years. My advice? You should find a Sigrid.

I think the right motivation is to do something you care about and you're good at. If you're not good at something, go find something. When you like something, it doesn't feel like work.

Specifically for young people pursuing careers, software is an endless frontier. There isn't any limit to what you can do. I can confidently argue that maybe only 1% or 2% of the possible applications have already been implemented or thought of.

Where are you with your work on the interplanetary Internet? The interplanetary work was started at [NASA's Jet Propulsion Laboratory] in 1998. We wanted to standardize space communications. We quickly discovered that the basic TCP/IP protocols would work on the surface of a planet or in a spacecraft, but they don't work well when speaking on an interplanetary basis. The space between planets is so big. Between Earth and Mars, it takes 20 minutes one way and 40 minutes round-trip for a radio signal to propagate between planets. Forty minutes is way long.

Delay and disruption [tolerant] networking protocols are implemented and developed. They're available in open source and being tested in near space. In deep space, you have these unavoidable, long speed-of-light delays. Tolerance for uncertain delays that cause disruption is exactly what's needed.

We're hoping to put tests together that will take us to Mars in the next decade. The intention is to standardize protocols so the same thing that allows you to take an arbitrary laptop and plug it in and talk would apply in deep space.

Join the discussion
Be the first to comment on this article. Our Commenting Policies