In Remembrance: Paul Baran, 84
Baran was a coinventor of packet switching, a serial entrepreneur, and a forefather of the Internet
Steven Cherry: Hi, this is Steven Cherry for IEEE Spectrum’s "This Week in Technology."
One of the forefathers of the Internet died this week—Paul Baran, at age 84. In 1959, Baran was graduated from UCLA with a master’s degree in engineering and joined the RAND Corporation, one of the few think tanks in existence back then. He began work on two technical papers for the Air Force on designing communications networks that had enough redundancy to survive an attack.
It was no idle fear. In 1961, three microwave relay stations in Utah were blown up. The New York Times reported at the time that the attack caused widespread communications disruptions and that a national defense circuit automatically shifted to an alternate route. Four members of a group that called itself the American Republican Army were eventually arrested.
Baran and his colleagues at RAND would eventually write a total of 13 papers on network survivability. The central concept was to digitize analog voice signals and turn them into data packets, which could go through one telecommunications switch or another as conditions in the network changed. This idea of packet switching was in the air; at exactly the same time, Donald Davies, in the UK, was working on it—he’s even credited with the term "packet switching"—and Len Kleinrock, then a graduate student at UCLA, published his thesis on the same idea in 1962 and as a book in 1964. They were the first published works on packet switching; those first two RAND papers were classified documents.
Baran aggressively pursued his ideas, both with AT&T—at the time the monopoly phone company in the U.S.—and at the Department of Defense. He didn’t have any direct success. But his RAND papers were studiously read by the designers of the ARPANET, the network that started in 1969 and grew up to eventually become the Internet.
My guest today is Robert W. Lucky. He’s an IEEE Fellow and an early winner of the Marconi Prize, which was awarded for his invention of the adaptive equalizer, the basis for the high-speed modems that connected the early message processors at the heart of the Internet and eventually allowed ordinary homeowners to join the Internet as well. He eventually became the head of Bell Labs, at AT&T, and later at Telecordia. Most important to Spectrum readers, for almost three decades he’s written an enormously popular column in the magazine called Reflections. He was also a close friend to Paul Baran. He joins us by phone from Fair Haven, N.J.
Bob, it’s a pleasure to edit your column, and it’s a pleasure to welcome you to the podcast, unfortunately for a sad occasion.
Robert W. Lucky: Well, thanks, Steven. It’s unfortunately a sad circumstance that we get together here.
Steven Cherry: Bob, I called Baran one of the forefathers of the Internet. I tried to use that word carefully. Some people use the word "pioneer," but that’s really the guys that built the ARPANET, guys like Kleinrock and Larry Roberts and Bob Cohn. Baran was of a slightly prior generation, I guess; he took the very first arrows in the back for some pretty revolutionary ideas. Do you think that’s fair to say? And maybe you could tell us what you take his achievements to have been and then maybe put them in perspective for us.
Robert W. Lucky: Well, yes, I definitely think he was one of the forefathers of the Internet. And you didn’t mention Vint Cerf, but let’s throw him in there, too, with Bob Cohn. Paul—I was one of the people you mentioned as studiously reading his paper from the RAND report that introduced me to packet switching many years ago, and definitely a beautiful, beautiful paper. So I think it’s very fair to call him one of the forefathers of the Internet. But you know, more than that I’d like to say Paul was just one of the greatest persons that I have ever known. And I could go into that some more—it—just his achievements belie his demeanor. You know he’s such an unassuming, brilliant giant that he just never wants to get credit for things; he always wants to be in the background and never step into the spotlight himself. And his generosity has been tremendous and just a fantastic person above what he’s done for technology. So it was a real pleasure to have known him for all these years, and I’m really greatly saddened by his passing this week.
Steven Cherry: You know, Bob, sometimes people are just in the right place at the right time, but he really pursued this idea of packet switching even when people weren’t very interested in it. Is that right?
Robert W. Lucky: Yes it is. I mean, like I was one of the people in communications research at Bell Labs, and I read his paper and I was really entranced by it. But we had spent a lifetime building up this circuit-switch network, and you know there was a fortune that had gone into building up around the entire world a circuit-switch network. And the thought sort of crossed my mind at the time, if we could do it all over again, would we build a packet-switch network? I probably didn’t think so, and it’s amazing how it turned out in the end that packet switching really triumphed. It’s a tremendous story of technology. Because I think the local area networks were the first place that you could really do this because you could build them from scratch. And then the ARPANET—when you got a chance to build a subnetwork, you know, from scratch, then that was a natural choice for them, too. But then also I think it was in the ’60s that Baran wrote his packet-switch paper. We didn’t really worry that much about data; data transmission was a small thing on the telephone network at that time. So there was never the thought that we could rebuild the telephone network for this particular method. That was probably good for data but was not good for voice.
Steven Cherry: Yeah, you know, I came across a quote that I’d like to read to you. You know I said before that AT&T wasn’t really interested—that’s kind of understating it—according to this quote, they didn’t really think it was possible, and I guess it’s because of this mindset toward the voice network. Let me just, um, this is from an interview that Baran did with Wired magazine about 10 years ago, and I think this quote is kind of priceless. So let me just read it, and we’ll get your take on it. These are Baran’s words: "The response was most interesting. The story I tell is of the time I went over to AT&T headquarters—one of many, many times—and there’s a group of old graybeards. I start describing how this works. One stops me and says, ‘Wait a minute, son. Are you trying to tell us that you open the switch up in the middle of the conversation?’ I say, ‘Yes.’ His eyeballs roll as he looks at his associates and shakes his head. We just weren’t on the same wavelength."
Robert W. Lucky: Well, you have to remember, Steven, that switching was done by relays then [laughs] and even after ARPANET came about, it took a lot of thought to actually tweak the algorithms so you could do voice over the packet-switch network. It wasn’t a natural fit, because voice is a rather continuous phenomenon, whereas data can be broken up very easily. And so to get the real-time nature of voice over packet switching, it wasn’t really straightforward. Now, of course, we do it very easily, but you know in the early days, it wasn’t that easy.
Steven Cherry: And it wasn’t even easy at the beginning of the ARPANET. That was 1969, and it wasn’t until the early ’80s that the Internet Protocol came to sort of be what it is now.
Robert W. Lucky: Yeah, VoIP is a natural thing today, but that only evolved in recent years. But packet switching for data—particularly for the military kind of communications that needed a lot of redundancy multipass phenomena that Baran had in mind when he wrote his report—I think it was very well suited to that. If you go to talk to AT&T people who made all of their money on voice in those days, I can see why they rolled their eyes. But their eyes didn’t continue to roll over the decades.
Steven Cherry: So Baran’s most lasting achievement, I guess, dates to the 1960s. But what else would you like us to remember about him?
Robert W. Lucky: You know, he started a number of companies in the cable business. For quite a few years I had a cable modem here that was made by one of his companies. And a lot of them were rather successful, particularly in the cable business, and even in recent years he’s described to me ideas he had for companies—for transmitting television and other things. Very good ideas, but I tell you, always for Paul, just unassuming. Brilliant guy, but totally unassuming. And as I said, I’ll just never forget his generosity. You mentioned the Marconi prize and that he won that, and he never wanted it even known that he contributed so much to the support of that society.
Steven Cherry: I’m told that he retired years ago and his friends could hardly tell the difference.
Robert W. Lucky: [laughs] That’s probably true, yeah. You wouldn’t know it, yeah. But he’s been in business with his son even in these companies, so he’s always been a technical thinker. And also Paul was a real hands-on engineer, and I admired him for that, too. You know a lot of people are just theoretical, but if you talk to Paul about ham antennas or something like that propagation something, his eyes would light up and he would tell you tales that would go into not just the physics of things but, you know, how things really worked.
Steven Cherry: So he was a real engineer’s engineer.
Robert W. Lucky: A real engineer. We value that—it’s almost a forgotten thing these days.
Steven Cherry: Very good. Well, thanks a lot for sharing us your remembrances of your friend.
Robert W. Lucky: It’s a real pleasure to talk to you about this, and as I said, it’s a sad occasion that we do talk about this because I really miss Paul.
Steven Cherry: We’ll bring you back to the podcast on a happier occasion, Bob.
Robert W. Lucky: Okay, great. Thanks, Steven.
Steven Cherry: Thanks a lot.
Robert W. Lucky: Bye-bye.
Steven Cherry: We’ve been speaking with telecommunications engineer, IEEE Fellow, and Spectrum columnist Robert W. Lucky about the enormous contributions of one of the forefathers of the Internet, Paul Baran, who passed away this week at age 84. For IEEE Spectrum’s "This Week in Technology," I’m Steven Cherry.
This interview was recorded 29 March 2011.
Audio engineer: Francesco Ferorelli
Follow us on Twitter @spectrumpodcast