1) The Internet : On its International Origins and its Collaborative Vision by Ronda Hauben rh120@columbia.edu 2) "It is a fair guess that textbooks of the next century will devote considerable attention to the Internet and larger changes in information and communication technologies that have emerged so dramatically in recent years." Roy Rosensweig American Historical Review December 1998 3) How Will the History Be Written? 4) Purpose of Talk 1. To distinguish between the ARPANET and the Internet 2. To document the international collaboration to create and develop the Internet 3. To consider the vision that inspired the develop of the Internet 5) Packet Switching ARPANET National Physics Laboratory (NPL) Cyclades 6) How could the Packet Switching Networks be connected? How could communication be possible across the boundaries of these dissimilar networks? 7) The Multiple Network Problem 8) 1970 - Discussion between American and British Research Groups How to link the US and UK networks together? They would utilize the connection between the US and Norway connection NORAD (NORwegian Seismic ARay) near Oslow to US. 9) Two problems: 1) Underestimated the tariff for extra drop off point 2) Timing from British National Perspective o Britain had just joined European Community o NPL and Department of Technology (UK) unable to take offer. o Had to concentrate on European initiatives like EIN (European Informatics Network.) o Peter Kirstein, a researcher at the University College London (UCL) was interested. 3) 1971 would attempt to set something up at UCL 10) Agreement reached between Larry Roberts at ARPA/IPTO and Peter Kirstein at UCL. 1) US to provide a TIP for the project - valued at 50,000 pounds 2) UK would be allowed to use the transatlantic link 3) UK would need to provide costs of breaking the communications link in London 4) By the end of 1971, technical proposal complete 11) 1972 - Kirstein trying to get support from the British government 1) By the end of 1962 the situation looked hopeless 2) Situation changed for Washington to NORSAR link o Scandanavian Tanum Earth Station in Sweden o U.S. - Norway connection no longer passed through the UK o New 9.6 kbps link between London and Kjeller would be needed o Cost would be even more expensive 12) Kirstein finds support 1) Two senior officials of the British Post Office o Murray Laver of National Data Processing Service o Alec Merriman of Advanced Technology 2) Donald Davies would sign personally for 5000 pounds 3) Kirstein tells Larry Roberts "We would proceed." 13) Kirstein still finds situation difficult Main British research initiatives in pursuit of X.25 14) IPTO invited Norwegian researchers to collaborate 1) Invitation to NTA (Norwegian Telecommunications Administration) 2) Invitation to Norwegian Defence Research Establishment (NDRE, "Forsvarets Forskningsinstitutt") 3) NDRE's interest in basic computing and networking research reason for Norwegian collaboration with IPTO - Yngvar Lundh 15) September 18, 1972 - Larry Roberts and Robert (Bob) Kahn visit Norway. 1) Meeting with Yngvar Lundh, research engineer at NDRE Finn Lied, director of NDRE and Karl Holberg, research superintendent of NDRE electronics department. 2) Meeting held in Oslo at civilian research administrative office (Royal Norwegian Council for Scientific and Industrial Research.) 3) Representatives present from other organizations in Norway. 16) Kahn and Roberts invite Norwegian researchers to attend ICCC'72 in Washington DC in October 1972. Lundh accepts invitation - Attends ICCC'72 on Octo 25 and 26, 1972 At ARPANET Demonstration Lundh invited to meeting at COMSAT Corporation Meeting discussion of reasons "for establishing a net of nets where each individual net would use the best low level protocol" for communication. Kirstein also attended, probably a number of US researchers from IPTO and researchers from other countries including UK and France. 17) Lundh built a small research group in Norway. Difficult to raise money and resources. 18) Norwegian ARPANET TIP put at NORSAR. not at NDRE TIP at civilian facility providing for more widespread participation. 19) For the Norwegian researchers 1) Hard to build research team when lack of funding. 2) Lundh believed the possibilities of "resource sharing" were fanatastic. 3) Resources sharing of expensive programs, special data, ideas, people with various interests and capabilities. 20) After ICCC'72 conference Bob Kahn left BBN to go to IPTO 1) Took over responsibility for Satnet project - already funded. to create a satellite-based packet switching network. (SATNET) 2) Began new initiative - create a ground based packet radio network (PRNET) 3) Kahn's Objectives: o Create multinode ground packet radio network each node would be mobile o Create a packet satellite network - utilizing INTELSAT satellites. o Two of networks would use radio transmission Third network - ARPANET used shared point to point leased lines from the telephone company. o Need for architectural conception to make resources sharing possible across boundaries Peer to peer relationship - open architecture 21) Spring 1973 - Kahn invited Vinton Cerf to collaborate with him INWG note #39 prepared for meeting at Brighton, UK September 1973 22) First presentation of draft paper describing philosophy and design for the TCP/IP protocol. "A Protocol for Packet Network Intercommunication" May 1974 IEEE Transactions on Communications 23) Need to write specifications to implement the design Cerf - two versions of specifications developed at Stanford U o December 1974 o March 1977 24) Research work done by US researchers and graduate students, several researchers from the UK, French researchers and researchers from Norway 25) Dag Belsnes - 1973 one year grant Norwegian Research Council Met Vint Cert at conference in UK in 1973. Accepted to be part of Digital System Laboratory at Stanford U "I got the opportunity to participate in his Protocol Design Group thatworked on creating a specification for the Internet Transmission Control Program. 26) Paal Spilling, a Norwegian researcher who would work with Yngvar Lundh, writes: "Although the TCP was specified in detail, it had to be considered a first approach towards making a reliable process-to-process communication tool in an internetwork environment...The results obtained, helped in the debugging of this first version of TCP, and uncovered some deficiencies in its design." 27) June 1973 TIP installed at Kjeller, Norway for NDRE researchers July 1973, UCL TIP in UK passing packets between US and UK Packets went from US via satellite to Tanum Earth Station in Sweden, via land and underwater lines to NORSAR in Kjeller, Norway, and then to London in the UK. (see diagram) 28) Kirstein describes: An early protocol experiment in late 1974 between Stanford and UCL. Spilling visiting UCL from NDRE and Judy Estrin, graduate student working with Vint Cerf at Stanford. 29) Spilling writes: "As I remember the fellows at the Stanford side may have been Judy Estrin and Jim Mathis. At the UCL side Frank Deignan, Andrew Hinchley and me. Frank was the implementer. 30) "It was extremely exciting to observe packets coming from Stanford and after an initial debugging being accepted and processed by Frank's implementation of TCP." Spilling 31) "One critical problem I can remember was that the TCP checksum was applied slightly differently at Stanford and then at UCL." Spilling 32) Invitation to Norwegian Telecommunications Adminstration Research Establishment (NTA-RA) led to "the free loan for experimental purposes of a spare channel in the INTELSAT IV satellite and a spare line between NDRE and the existing Scandinavian Satellite Earth Station at Tanum, Sweden." Lundh 33) Commercial traffic was prohibited in the ARPANET from the outset and that was still true as the Internet was developed.