It is one of the quirks of science that the same idea can arise simultaneously in different places. Who gets the subsequent credit often depends on who shouts loudest.
So it is with packet switching. This is a way to break up messages into numbered packets that can be addressed and “posted” on a data network. Thanks to the addressing, each packet can take a different route through the network to its destination. On arrival at their destination, the packets are reassembled in the correct order and the contents delivered.
In the 1960s, three teams, two in the US and one in the UK, were working independently on packet switching, unaware of each others’ efforts. It was not until Donald Davies and Roger Scantlebury of the UK’s National Physical Laboratory (NPL) presented a paper on packet switching concepts at a 1967 conference in Gatlinburg, Tennessee that the penny dropped.
The Americans quickly adopted Davies’ ideas on packet switching for Arpanet, the new national network designed to preserve the military’s command and control communications system following a nuclear attack. Backed by Pentagon money, Arpanet grew quickly to include research institutes, universities and private sector firms, especially those that supplied goods and services to the military.
In the fullness of time Arpanet became the internet, and the packet switching technology developed at the NPL became and remains the primary means of moving voice, data, and images around the world.
While the Americans went on to build giant firms such as Cisco and Juniper Networks on the back of packet switching, Davies and his team went back to their lab, job done.
Speaking at the National Museum of Computing, which has opened a special gallery to commemorate the NPL’s contribution to the birth of the internet, Brian Aldous, one of the technical experts on Davies’ packet switching development team, tells Computer Weekly’s Ian Grant how it happened.