Europa heeft een verhaal te vertellen over internet: audio-bestand (en) - Hoofdinhoud

[The last of my guest blogs for the summer, by Mariann Unterluggauer, free, ("but not as in 'free beer'") European journalist ]

Forty years ago, in October 1972, a small team of young American researchers presented at the Hilton hotel in Washington DC the “Arpanet“. Also around was the British computer scientist Derek Barber. He presented at the International Computer Communications Conference a European research project called “COST Project 11". COST 11 got renamed to “European Informatics Network” (EIN), “because it sounded better”.

Enjoy a radio play of a nearly forgotten story about the early days of European networking. In fact, we should be proud of it, study it, celebrate it, and for once, learn from it, as some sound snippets don’t sound outdated and might well trigger a feeling of déjà-vu, especially for those who are aiming towards next generation Internet.

Consider: What problem did they try to solve in the early days? What solutions did they come up with and why? What was spoiled by whom and why, and what makes the difference between the US and Europe in developing computer networks? These questions are answered, at least partly, in my short radio story, but pursued further they can help to find out what parameters are still at place in Europe and shape the design ideas for the next Internet.

A honest and critical look over these pioneer’s shoulders can be of help for users and journalists, in order to decipher marketing hype, for start-ups and investors in order to find real beef instead of copycats, and for politicians in order to find a balance between market interests, benefits for society and to foster - dare I say it - European pride. Networking history, imho, should be considered as part of our cultural heritage.

Soundfile: European networking - the beginning (6")

Credits, in order of first appearance:

Peter Kirstein: University College London. He set up the link to the Arpanet in the UK in 1973. Since then he has been involved in setting up global Internet operation in the Southern Caucasus and Central Asia regions. (Interview 2003)

Louis Pouzin: Built the French network Cyclades (1972-1978), and developed Datagrams further (used for example for VOIP services as Skype). Today he works on internet governance and an open root project for the Savoir-Faire company (Interviews 2008, 2012)

Leonard Kleinrock: pioneered the mathematical theory of packet networks. He transported many of those early Internet pioneers with his sweet 1963 Chevy Impala Super Sport Convertible, all over Southern California. (Interview. 2012)

Ben Segal: CERN IT Department, retired. Was mentor of Tim Berners-Lee; today he is involved in the LHC++@Home, a crowd-sourcing project at CERN. (Interview 2009)

Maurice Allègre: From 1968 to 1974 in charge of French government’s policy for the development of the computer industry, and the use of computers in society. He was leading the French Délégation à l’Informatique in Brussels in the 1970s. (Interview 2008)

Tomaz Kalin: Is currently serving on the Board of Directors of Telecom Slovenia. He was a member of the Management Committee of COST 11 Program. (Interview 2009)

Larry Landweber: has been involved in the development of the international Internet. In 1979, he proposed CSNET, was senior advisor to the US National Science Foundation and helped to bring GENI alive, the US version of Future Internet endeavours. (Interview 2007)

Charles Herzfeld: became ARPA/DARPA Chief 1965 - 1967. He signed the first check for the Arpanet in 1966 to support Robert Taylor to make J.C.R. Lickliders dream come true. (Interview 2006)

Bob Kahn: He was responsible for the system design of the Arpanet and wrote together with Vint Cerf the TCP specification in 1974. Developed later on the Handle System, a general purpose distributed information system at CNRI. (Interview 2004)

Elisabeth Feinler (Jake) managed documentations for the ARPANET, and the WHOIS server. Doug Engelbart brought the trained bio-chemist to computer networking. (Interview 2012)

Vint Cerf: Wrote together with Bob Kahn the first proposal for the internetworking protocol TCP. Today he works as Internet evangelist for Google. (Interview 2012)

Acronyms:

CCITT: today ITU

PTT: PTT stands for post, telegraph and telephone, state monopoly

Early European Networks, mentioned in the play

Arpanet: USA1969 -1983

EIN: European Informatics Network: officially from 1971 - 1978

Cyclades: packet switching network, 1972 - 1978 (but in use, illegally, a little bit longer, who knows.)

NPL- Net: A Campus network developed in 1969 under the lead of Donald Watts Davies at the National Physical Laboratory, near London. Donald W. Davies started to work on packet switching, a term he invented, in 1965. He presented his concept to the standardisation body CCITT (today ITU) in 1967, as well as to his US-colleagues.

Transpac: 1978 - France Telecom network

Addendum: Arpanet and European Informatics network, seen in parallel

1967: At the same time as feasibility studies for the ARPANET were done in the US, the proposal for EIN reached the EEC.

1968: The plan to build an experimental, pan-European Informatics network (EIN, COST Project 11) gets accepted.

1969: USA, the first hosts are connected to the Arpanet

1970: It took the EC two more years to decide who should take the technical lead of project EIN. Finally it went to the UK to Derek Barber. Nobody was more surprised as the British computer scientist himself. English was not an official EEC language, and some years before Charles de Gaulle had prevented once more the attempt of the UK to join the club. (de Gaulle favoured an expansion to the East instead.)

1971: the international UN treaty was presented and signed partially in November 1971, by eleven players. (COST was created as a scientific program so also non-EEC members can join, if they pay) (LINK: TREATY pdf. see attachments)

1972: First public demonstration of the ARPANET. Louis Pouzin starts work on Cyclades.

1974: EIN Specifications agreed upon. Contracts signed to Logica (UK), SESA (France), Selenia (Italy) and FIDES (Switzerland) to connect, and to equip the five nodes (Milan, Ispra, Zurich, Paris, and London). In the USA, Bob Kahn and Vint Cerf made the first step towards the Internet, and present their idea for an Internetworking protocol TCP. At that time the name Internet didn’t exist, in the handbooks one can find the term “Catenet” instead. Derived from catena, and invented by Louis Pouzin.

1976: The EIN sub-network began to operate, and the project got a deadline: The Commission agreed, that the PTTs should take over, and build their own network called EURONET.

1978: Final public presentation of EIN in Paris. Only a small brochure of this European episode with the subtitle “concourse of computers” survived in private archives. 35 years ago EIN as well as the French Network Cyclades was brought to an end. And as Larry Landweber says in the radio play: This is something that never happened to the US experimental research network Internet

1983: The evaluation report of the EC on “EIN” published. The evaluators put two and two together and stated: The project had been successful in spreading and coordinating research in Europe, but the industry was not interested and the PTTs saw it as a potential menace. (PTT stands for post, telephone and telegraph, their monopoly ended officially, and thanks to the EU, in 1998.)

In the USA the networking crowd turned the lever and switched ARPANET to Internet.

Worth to muse on

At the end of 1960 two European groups had real knowledge of packet switching networks: The group of Donald W. Davies, who had built the NPL network in the UK (and of which Derek Barber was a member), and the team around Louis Pouzin, who had built the French network Cyclades. Pouzin developed the concept of “Datagrams” further, which was implemented in the Internet Protocol Suite in the 1970s (still in use today), but delayed from being standardized in Europe until 1994. Other members of EIN have been young assistants, who studied proceedings of the ARPANET, but had no hands-on experience, yet stars in their eyes.

“Internetworking” was not considered a career path at all. Europe as well as the US only had a small group of dedicated people each, who “fitted into one smallish room”. The European group fought traditional telecom paradigms, and tried to convince regulators that Europe, for once, needed to act as united. Instead, the PTTs, vendors, and governments headed for OSI, “Open Systems Interconnection”. They even gave those European researchers a hard time, who dared to experiment with Internet Protocol Suite TCP/IP in the 1980s. But then came the web, and again a small crowd made a difference …

In order to make one step forward, the midterm evaluation of the Digital Agenda should not circle around the question whether or not its visions can be delivered on time (no vision ever has), and pictured in scorebooks. It’s time to analyze who and what spoils the process. It’s time to add human values, social effects and needs to the equation, not just processor power, markets and old clubs’ interest. This can make the difference between old and new networking paradigms.