As Paul Baran passed away this week, we will miss his great inventiveness as well as his warm, humble nature.
Paul was one of the three people who conceptualized packet switching in the early 1960s. There were Len Kleinrock at MIT, Donald Davies in the UK, and Paul at Rand. When I undertook to build the ARPANET in 1967 at Advanced Research Projects Agency (ARPA), I had the advantage of Kleinrock’s book on networks and queuing theory, Davies’s work on packet switching, and Paul’s 11 papers on a secure, reliable command and control network.
Three years earlier, Paul had attempted to get several government agencies to consider his network concept, but he had received no support. He must have been very frustrated to have a great idea but to meet a brick wall. Fortunately for us all, ARPA took the risk; and based on my earlier success with a cross-country packet communication test and Kleinrock’s well proven queuing theory, the technology was finally in hand to move ahead with the first packet network, the ARPANET.
Again, we must recognize Paul for laying the conceptual groundwork.
Paul, however, did not stop working on new ideas and continued to start five companies that expanded the utility of packet switching. Metricom and Ricochet were wireless packet concepts; Com21 built cable modems; GoBackTV was into IPTV; and Plaster Networks provided home and small office networking. His mind continued to generate new ideas throughout his life.
Paul was always lighthearted and never was one to look to claim credit. In his 1999 speech to the CableLabs winter conference he stated: “On occasion, I get accused of creating the notion of the ARPANET. Not guilty!”
What Paul did do as the ARPANET evolved into the Internet was to proceed to use his tremendous inventiveness to extend the Internet to packet wireless and packet cable technology, among other things.
Paul played a strong continuing role in the development of packet technology. We will all miss him.
— Dr. Lawrence G. Roberts, Chairman, Anagran Inc.