Launching an interplanetary Web

Vint Cerf, known as the father of the Internet, is working on a proposal to create a network of Internets between planets, satellites, asteroids and robotic spacecraft.
Written by Steve Kovsky, Contributor on
Vint Cerf, known as the father of the Internet, has got his eyes on the stars. The senior vice president for Internet architecture and technology at telecommunications company WorldCom is working on a proposal to create a network of Internets to facilitate communication between planets, satellites, asteroids, robotic spacecraft and crewed vehicles.

While it may sound like the stuff of science fiction, Cerf's participation lends a certain gravity to plans for the stellar Net system. As one of the co-creators of the TCP/IP protocol and basic architecture of what we now know as the Internet, Cerf has been at the forefront of the Net revolution. In 1997, both he and scientific partner Robert E. Kahn were recognized for their work when President Bill Clinton presented them with the US National Medal of Technology.

In a recent interview, Cerf talked about plans for an interplanetary Internet, the dot-com malaise, and whether Web-surfing refrigerators are really something we need.

Q: Can you tell me about your project to extend the Internet into outer space or rebuild the Net so that it can support the demands of planetary exploration?
A: The Internet works essentially pretty well on the surface of our planet Earth...and it should work perfectly well on the surface of other planets or satellites, or even space vehicles that are out there in the solar system. So the basic Internet protocols will work just fine in many different parts of the solar system.

The only issue that really comes up is when you go interplanetary--when you have to carry information from one planet to another. And there you get into significant delays because of the astronomical distances involved. You may run into serious variations in error rates. You may run into differences in data rates when you send or receive data.

All of those variations caused us to design a new set of protocols, in addition to the ones that are normally part of the Internet, to work on an interplanetary basis. So that's what I've been spending the last two and a half or so...almost three years now with the Jet Propulsion Lab engineers to develop.

So why do we need the Internet not only to work well on other planets, but to work between planets? Is there demand?
Oh, well...you haven't talked to my friends, the Martians! (Laughs.) In fact, we have lots of robots out there. And I don't mean things with legs crawling around, necessarily. What we have are space vehicles with sensing equipment on them, and they are often on missions to various parts of the solar system, and they need to relay back information to us that we can analyze what they've discovered. It's partly for that reason that we're interested in building a standardized protocol architecture for interplanetary communication, so when we have space vehicles that are either landing on the surface of the planet or possibly proliferating across the planet sensor information that could be linked up to an orbiting satellite by radio, that we have an architecture that will support these interplanetary transmissions.

So our focus of attention is on interplanetary for purposes of communicating with robotic devices, sensors and analyzing. In the longer term...2018 and out, there is a reasonable possibility that we will see some manned missions that might take us beyond the orbit of the moon. But that hasn't been the primary motivation. It's been primarily in support of space exploration using small vehicles.

When you created TCP/IP, essentially giving life to what we know now as the Internet, I would guess you never saw the concept being applied to space exploration. What else has surprised you about recent innovations?
Well, the ones that really get me, of course, are the ones where you have these Internet-enabled appliances; things like refrigerators would have Internet capability and maybe a scanner that can sense bar codes and things like that. So now you have a refrigerator that knows what's in it, and it's on the Net. So when you come home, you discover it's been surfing the Net, finding recipes for things it can make with what it knows it has inside.

Or worse, you have bathroom scales that are Internet-enabled. Then you get on the scale and it sends your weight to the doctor, and that becomes part of your medical record. And then one day you come home and you discover diet recipes on the refrigerator because it has got the same information that was sent to the doctor and decides you need to go on a diet.

In fact, the most recent nightmare is the nano-engine guys who go off and build these tiny little Internet-enabled radio transceivers, and they get embedded in your clothes. So you interrogate your sock drawer and you get back a message saying: "Sock drawer contains 17 matched pairs of socks and there are three unmatched socks. Sock #114327L is missing." And so you send out a broadcast around the house and sure enough, sock #114327L responds: "I'm in the house at this latitude and longitude behind the sofa." So now you have a way of solving the lost sock problem. I thought that was cool.

How about this: Can it tell me if the socks I'm wearing don't match, before I go out of the house?
There you go! Absolutely. Sure. Because they both have embedded in them knowledge about what color they are, so when you put them both on, and if they're signaling to each other, you could get an alarm going off saying: "Your feet don't match."

Now, there is a downside to this. If your clothes can be located, you can imagine calling home and saying: "Hi, I'm going to work late in the office tonight." You get this puzzled response saying: "Well gee, that's really interesting because your shirt is down on 15th Street at the bar!"

That could pose some problems...
Yeah, it could be problematic. So sometimes all these amazing ideas have their downsides. But I am convinced that we will have literally billions of Internet-enabled devices. I have one sitting on my desk today. It's an Internet-enabled picture frame. It does one thing--it plugs into the wall, gets power from the telephone service, then it dials up the Net and it downloads 10 pictures from a predetermined Web site, which I can populate, and then it just rotates through the 10 pictures. So it's great for the grandparents. You stick it on the desk, and they just turn it on. And every so often you go and update the set of pictures that that machine is going to display, and it downloads them and captures them and then displays them so they can watch the grandchildren growing up without actually having to be there. That is an absolutely reasonable application for today, and the device costs around US$200.

We've seen so many of these applications being launched in the last year or two, and so many of them have fallen flat, particularly in the pure-play dot-com business.
The "dot-com done dot-bombed". Right.

What's your perspective on that? I've always perceived you as much more of a scientist than a businessman.
I certainly wouldn't run around telling everybody I knew much about running a business, but I do have an opinion! I actually have two opinions. The first one is that most new businesses fail. It has little to do with dot-com, dot-bomb or anything else. It has to do with new businesses whose business models didn't quite work quite the way the founder hoped. So we're just getting a spotlight on a lot of businesses that have failed in the past...not-too-long-ago past, as they're Net-related and because we've been focused on that particular segment of our business environment.

They were also funded at a much higher, more visible level than the average mom-and-pop store.
In many cases, that's absolutely correct. So that gave them the added visibility that normal start-ups wouldn't have had. And so their failures are that much more spectacular. But you know, I have this other view of it: It's wrapped around a metaphor that says that this kind of experience in the market is a lot like a forest fire. There was a time when we thought forest fires were really bad things, and we tried very, very hard to prevent them or to put them out once they started. But then some of the scientist-ecologists were telling us: "No, no...wait a minute. Actually a lot of forest fires are good." I don't mean having a lot of them is good, but many of them are good because they burn away the underbrush; they burn away the dead trees; they create space and nutrients for new growth.

And that may be what's happening to us in the Internet world, where a lot of these essentially dead business ideas are burning up and blowing away, but they're leaving behind some room for capital growth or capital investment in companies whose business plans actually make sense.

So in a way this whole thing is painful as hell; many people got hurt. But it's probably very healthy to finally get people focused on much more care analyzing business models and developing them and making sure they have credibility before you're going to get any money.

Did you lose any money personally?
Oh, like a lot of people, I have some paper losses, but I'm not about to jump out of the 12th-floor window or anything. For one thing, the market has always had its ups and downs. Over the 40-some years I've invested in the market, it's returned an average that's pretty steady--it's somewhere between 8 percent and 12 percent. And that's not bad. You could live with that. It's sort of like Einstein--remember when they asked him, "What's the most powerful force in the universe?", and he waited for a moment, and he looked at the guy and he said, "Well, it's compound interest."

So even though this recovery has been painful, and it's not over by any stretch, this too will pass, as have many of the other crises in the market. And so I'm not discouraged at all. We've had a good, solid dose of reality.

Editorial standards