Internet Buzz with Richard Wiggins | 2

East Lansing, Michigan
An Interview with Vint Cerf

By Richard Wiggins

ecently Dr. Vinton Cerf gave the keynote address at a conference on the information revolution. Cerf is known as "the father of the Internet" because of his key role in developing the Internet Protocol. Currently Cerf is a vice president in the Internet division of telecommunications giant MCI. (Like everything else in the Internet business, the future of MCI's Internet division is in flux; MCI is working to spin off the division to satisfy antitrust concerns arising from the Worldcom merger.)

Prior to his keynote address to the conference, which was sponsored by the University of Michigan on behalf of Marshall Scholars, Dr. Cerf answered questions in a press conference. With him was Dr. Douglas Van Houweling, who leads the Internet2 and Abilene projects. Your intrepid WebReference reporters were lucky enough to ask Dr. Cerf a number of questions about the Internet. Following is Part 1 of our interview; see also Part 2 of our interview with Vint Cerf.

What comments do you have on the reliability and quality of service of the Internet today? Are things getting worse, or are they getting better?

Actually, that's a question that both of us ought to address. First of all, it has to get better. I don't mean to say that just because you said it has to get better, then it will, but people are relying on the Internet more and more. The research community that Doug is with is treating this network as a tool that they expect to be available, and running, and at very high speeds. We see the general public becoming increasingly dependent on the Internet, and so you'll hear the Internet engineers … at least in our shop … making the same kinds of noises that the telephone engineers made years ago – which is that well, we have to have dual power supplies, we have to make sure we have automatic backup always available, and all these other things … from the fiber level on up. My sense is that we will do that. It's certainly achievable. And it's exciting to think that the technology is reaching the point where we have to make reliability every bit as important as any other performance metric.

One thing that's clear to me is that the Internet is today a good deal more reliable than it was 9 months ago. I think everybody has seen a discernible increase in quality in the Internet that we all use. I think that's an enormous compliment to the guys that you just were talking about, who've been battling growth at 15 to 20% a month, and they're still managing to make it work a little better every month.

So I'm optimistic. But the other side of that same coin is that nobody really trusts the Internet yet with their mission-critical work. It's not that reliable. And, for higher education – especially the research community – it's now becoming mission-critical. And we have no other way to do it; we have to make the Internet work at that level, which is a lot of what motivates Internet2.

Of course, we get certain experience directly from Doug and his colleagues, because MCI runs something called …vBNS which is one of the networks used by that community to use the high-bandwidth, Internet2 kind of services. We've discovered that the academic world, which used to be relatively patient about outages and delays and things of that sort has become increasingly impatient for precisely the reason that they're relying on this stuff.

More...