Paul Baran was an engineer at the RAND Corporation whose early-1960s work on survivable communications laid one of the two independent foundations of packet switching. Confronted by the question of how a command-and-control communications system could keep working after a nuclear attack destroyed much of it, Baran concluded that the answer was a distributed network: a mesh of many nodes with redundant links, in which there is no critical center whose loss disables the whole.
His ideas were published by RAND in 1964 as the eleven-volume series “On Distributed Communications.” The first memorandum (RM-3420, “Introduction to Distributed Communications Networks”) set out the system concept, arguing for redundancy as the means of building networks that could withstand heavy enemy attack. Crucially, Baran proposed breaking messages into standardized “message blocks” that would be passed from node to node by a simple, adaptive “hot-potato” routing scheme, each node forwarding each block toward its destination without any central controller directing traffic.
As the Internet Hall of Fame summarizes, Baran “created a communications system that could survive the damage of a nuclear weapon,” initially calling the process “message blocks”; the British researcher Donald Davies independently reached similar conclusions and supplied the name that stuck, “packet switching.” The two had no knowledge of each other’s work until later, a striking case of simultaneous invention.
Baran graduated from Drexel University in 1949 and earned a master’s degree at UCLA. Beyond networking he founded the Institute for the Future and worked on wireless spectrum technologies, and he received the IEEE Alexander Graham Bell Medal and the National Medal of Technology and Innovation. He died on March 26, 2011, and was inducted into the Internet Hall of Fame in 2012. His RAND reports are regularly cited as a primary origin point for the architecture of the Internet.