You may have thought that the development of the internet was a purely US affair. You’d be wrong. Before we go too far, let’s clarify our terms. The UK’s Tim Berners-Lee devised the overwhelmingly popular application of the internet, the World Wide Web. But, not the internet itself – the network upon which the World Wide Web runs.
No less an authority than Vint Cerf, co-author of TCP/IP and co-founder of the institutions that administer the internet, says you’d be wrong too. Cerf addressed an online seminar run by Archives of IT and BCS for over 300 attendees on 6 January 2022 and focused on the UK contribution to early work. The contribution of people in the UK was ‘absolutely essential’ to the development of the internet, Cerf says. UK contributions from the UK were ‘extremely important’ to its development.
Many of the Archives’ 200 interviews and 30,000 pages of publications support Cerf’s assertion, with details of work done by British engineers and articles charting the growth of its use. The Archives show that the UK contributed the first implementation of a packet-switched network, the basis of TCP/IP; the first implementation in the world of TCP/IP; and the first international node of the internet outside the USA.
The UK’s contribution to the history of the internet
The idea of switching packets and not circuits was first outlined and implemented by Donald Davies at the National Physical Laboratory (NPL) in Teddington, Southwest London. He had joined NPL in 1947, and in the early 1960s was working on data communications, which he realized happened in bursts, quite unlike voice communications, which were at a steady rate.
Digital communications between computers were, therefore, typically over a line dedicated to a single connection which was then only used for a fraction of the available time. Better to split the data into ‘packets’, a term he chose after advice from a linguist, and send it over the line with other packets from other ‘conversations’. He outlined this in 1965 and the idea was floated by a colleague at a US conference in 1967 where it was taken up.
Alas, Davies died in 2000 before the Archives could capture his oral history, but another interviewee, Ann Moffatt, recalled a meeting with Davies. It was a special Computer Society talk, but held at the NPL and Donald Davies was working on what we later came to call packet switching, but we didn’t call it that then.
But the idea was to send packets of data across the world! Across England or across the world, using computers, sending files of data. On telephone lines. I mean this was just amazing that somebody was thinking about this, and I thought if we could merge together computers and telephone lines, gosh, that would open up just such an interesting life.’
Internet history and first implementations
And merge they did. TCP/IP was devised in the USA by Vint Cerf and Bob Kahn, having heard of Davies’ work at the 1967 US conference. Its first implementation in the world was in the UK at University College London (UCL) by Peter Kirstein CBE and colleagues.
Kirstein told the Archives. Bob was my project officer, and Vint was a fairly junior professor at Stanford. They thought up this new internet protocol, and one part of anything like that was, you had to have implementations. So, the first of three implementations done was by this junior academic at Stanford called Vint Cerf, somebody in Bolt, Beranek & Newman, who were the people who actually provided the computers for the ARPANET, and myself, at least one of my people, or two, several of my people. And we had the first actual implementations between the three of us. So, we were doing the research on the internet from ’75 or so onwards.
As Kirstein said: Around that time, ’76, ’77, there was a lot of standards work going on in Europe, and the British were part of that standardization work. They did not approve of the ARPANET [the precursor of the internet] because they regarded that as an experiment, and although the academics liked it, and they tolerated it, they certainly didn’t regard it as mainstream, and were perfectly happy to have me connecting in the ARPANET, but they didn’t want people working on the internet side of things. So I was actually ordered to stop working on the internet protocols.