TheInternational Network Working Group (INWG) was a group of prominentcomputer science researchers in the 1970s who studied and developedstandards andprotocols for interconnection ofcomputer networks. Set up in 1972 as an informal group to consider the technical issues involved in connecting different networks, its goal was to develop an international standard protocol forinternetworking. INWG became a subcommittee of theInternational Federation for Information Processing (IFIP) the following year. Concepts developed by members of the group contributed to theProtocol for Packet Network Intercommunication proposed byVint Cerf andBob Kahn in 1974 and theTransmission Control Protocol andInternet Protocol (TCP/IP) that emerged later.
The International Network Working Group was formed bySteve Crocker,Louis Pouzin,Donald Davies, andPeter Kirstein in June 1972 in Paris at a networking conference organised by Pouzin.[1][2] Crocker saw that it would be useful to have an international version of the Network Working Group, which developed theNetwork Control Program for theARPANET.[3]
At theInternational Conference on Computer Communication (ICCC) in Washington D.C. in October 1972,Vint Cerf was approved as INWG's Chair on Crocker's recommendation.[4][5][6][nb 1] The group included American researchers representing the ARPANET[nb 2] and theMerit network, the FrenchCYCLADES andRCP networks,[nb 3] and British teams working on theNPL network,EPSS, andEuropean Informatics Network.[4][7]
During early 1973, Pouzin arranged affiliation with theInternational Federation for Information Processing (IFIP). INWG became IFIP Working Group 1 under Technical Committee 6 (Data Communication) with the title "International Packet Switching for Computer Sharing" (WG6.1). This standing, although informal, enabled the group to provide technical input on packet networking toCCITT andISO.[4][6][8][9][10] Its purpose was to study and develop "international standard protocols for internetworking".[11]
INWG published a series numbered notes, some of which were alsoRfCs.[4][12]
The idea for arouter (called agateway at the time) was initially described in "INWG Note 1", a report written in October 1972 by Donald Davies (NPL, UK), P. Shanks (Post Office, UK),Frank Heart (BBN, US), B. Barker (BBN, US),Rémi Després (PTT, France), V. Detwiler (UBC, Canada) and O. Riml (Bell-Northern Research, Canada).[4][13] These gateway devices were different from most previous packet switching schemes in two ways. First, they connected dissimilar kinds of networks, such asserial lines andlocal area networks. Second, they wereconnectionless devices, which had no role in assuring that traffic was delivered reliably, leaving that function entirely to thehosts.[14] This particular idea, theend-to-end principle, had been pioneered in the CYCLADES network.[15]
A second sub-group considered host-to-host protocol requirements. This group consisted of Barry Wessler (ARPA, US), Vint Cerf (Stanford University, US), Kjell Samuelson (Stockholm University), Derek Barber (NPL, UK), C.D. Shephard (Deptartment of Communications, Canada), Louis Pouzin (IRIA, France), Brian Sexton (NPL, UK), William Clipsham (UK), Keith Sandum, Alex McKenzie (BBN, US), and Jeremy Tucker (Logica, UK). In their initial report in October 1972, they listed existing protocols for various networks that they planned to review, including the "Walden Message-Switching Protocol, ARPA H-H Protocol, NPL High-Level Protocol, CYCLADES Protocol, SPSS Protocol, etc."[4][16]
INWG met in New York in June 1973. Attendees included Cerf,Bob Kahn, Alex McKenzie,Bob Metcalfe,Roger Scantlebury,John Shoch andHubert Zimmermann, among others.[4][17][18][19] They discussed a first draft of anInternational Transmission Protocol (ITP).[4] Zimmermann and Metcalfe dominated the discussions; Zimmermann had been working with Pouzin on the CYCLADES network while Metclafe, Shoch and others atXerox PARC had been developing the idea ofEthernet and thePARC Universal Packet (PUP) for internetworking.[20][17] Notes from the meetings were recorded by Cerf and McKenzie, which was circulated after the meeting (INWG 28).[4][12] There was a follow-up meeting in July. Gerard LeLann and G. Grossman made contributions after the June meeting.[4]
Building on this work, in September 1973, Kahn and Cerf presented a paper,Host and Process Level Protocols for Internetwork Communication, at the next INWG meeting at theUniversity of Sussex in England (INWG 39).[21] Their ideas were refined further in long discussions with Davies, Scantlebury, Pouzin and Zimmerman.[22]
Pouzin circulated a paper onInterconnection of Packet Switching Networks in October 1973 (INWG 42),[4][12] in which he introduced the termcatenet for an interconnected network.[4][23] Zimmerman and Michel Elie wrote aProposed Standard Host-Host Protocol for Heterogenous Computer Networks: Transport Protocol in December 1973 (INWG 43).[12] Pouzin updated his paper withA Proposal for Interconnecting Packet Switching Networks in March 1974 (INWG 60),[12] published two months later in May.[24] Zimmerman and Elie circulated aStandard host-host protocol for heterogeneous computer networks in April 1974 (INWG 61).[12] Pouzin publishedAn integrated approach to network protocols in May 1975.[25]
Kahn and Cerf published a significantly updated and refined version of their proposal in May 1974,A Protocol for Packet Network Intercommunication.[4] A later version of the paper acknowledged several people including members of INWG and attendees at the June 1973 meeting.[26] It was updated in INWG 72/RFC 675 in December 1974 by Cerf, Yogen Dalal and Carl Sunshine, which introduced the terminternet as a shorthand forinternetwork.[27]
Two competing proposals had evolved,[29] the early Transmission Control Program (TCP), originally proposed by Kahn and Cerf, and the CYCLADES transport station (TS) protocol, proposed by Pouzin, Zimmermann and Elie. There were two sticking points: whether there should be fragmentation of datagrams (as in TCP) or standard-sized datagrams (as in TS); and whether the data flow was an undifferentiated stream or maintained the integrity of the units sent. These were not major differences. After "hot debate", McKenzie proposed a synthesis in December 1974,Internetwork Host-to-Host Protocol (INWG 74), which he refined the following year with Cerf, Scantlebury and Zimmerman (INWG 96).[4][20][28][30]
After reaching agreement with the wider group,[nb 4] aProposal for an international end to end protocol, was published by Cerf, McKenzie, Scantlebury, and Zimmermann in 1976.[31][32][33] It was presented to the CCITT and ISO by Derek Barber, who became INWG chair earlier that year.[4] Although the protocol was adopted by networks in Europe,[34] it was not adopted by the CCITT, ISO nor the ARPANET.[4]
The CCITT went on to adopt theX.25 standard in 1976, based onvirtual circuits.ARPA funded testing of TCP in 1975 at Stanford, BBN and University College London.[35] With funding from ARPA, another group published theInternet Experiment Notes from 1977 to 1982. This group developedTCP/IP, theInternet Protocol as connectionless layer on top of theTransmission Control Protocol as a reliable connection-oriented service, which reflects concepts in Pouzin's CYCLADES project.[36]
Ray Tomlinson proposed a network mail protocol inINWG Protocol note 2 (a separate series of INWG notes), in September 1974.[4] Derek Barber proposed a network mail protocol and implemented it on the European Informatics Network, which he reported in INWG 192 in February 1979.[37] His work was referenced byJon Postel in his first paper on Internet email, published in theInternet Experiment Note series.[38]
Alex McKenzie served as chair from 1979-1982.[4][11] A new international effort, beginning in 1978, led to theOSI model in 1984, of which many members of the INWG became advocates.[5][39] During theProtocol Wars of the late 1980s and early 1990s, engineers, organizations and nations became polarized over the issue of which standard, the OSI model or the Internet protocol suite would result in the best and most robust computer networks. ARPApartnerships with the telecommunication and computer industry in the 1980s led to private sector adoption of the Internet protocol suite as a communication protocol.[5][40][41]
McKenzie became the Secretary in 1983 and Carl Sunshine, who had worked with Vint Cerf and Yogen Dalal at Stanford on the first TCP specification, became the INWG chair until 1987, when Harry Rudin, at the IBM Zurich Research Laboratory, took over.[4][42]
The INWG continued to work on protocol design and formal specification until the 1990s when it disbanded as theInternet grew rapidly.[4] Nonetheless, issues with the Internet Protocol suite remain and alternatives have been proposed building on INWG ideas such asRecursive Internetwork Architecture.[28]
The work of INWG was significant in the creation of routers, the Transmission Control Program, and email which all ultimately became pivotal in the working of the Internet.[14][43]
... the International Network Working Group was created ... to draw a larger cohort of people into this whole question of how to design and build packet switch networks. That eventually led to the design of the Internet.
— Vint Cerf (2020)[44]
The group had about 100 members; the initial two sub-groups consisted of the following:[4][9][13][16]
But the ARPANET itself had now become an island, with no links to the other networks that had sprung up. By the early 1970s, researchers in France, the UK, and the U.S. began developing ways of connecting networks to each other, a process known as internetworking.
INWG#1: Report of Subgroup 1 on Communication System Requirements by Davies, Shanks, Heart, Barker, Despres, Detwiler, and Riml. They wrote: "It was agreed that interworkingbetween packet switching networks should not add complications to the hosts, considering that networks will probably be different and thus gatewaysbetween networks will be required. These gateways should be as uncomplicated as possible, whilst allowing as much freedom as possible for the design of individual networks". INWG#1 clarified that gateways and simplicity were accepted concepts when INWG was formed.
Roger Scantlebury was one of the major players. And Donald Davies who ran, at least he was superintendent of the information systems division or something like that. I absolutely had a lot of interaction with NPL at the time. They in fact came to the ICCC 72 and they had been coming to previous meetings of what is now called Datacomm. Its first incarnation was a long title having to do with the analysis and optimization of computer communication networks, or something like that. This started in late 1969, I think, was when the first meeting happened in Pine Hill, Georgia. I didn't go to that one, but I went to the next one that was at Stanford, I think. That's where I met Scantlebury, I believe, for the first time. Then I had a lot more interaction with him. I would come to the UK fairly regularly, partly for IFIP or INWG reasons
The term "catenet" was introduced by L. Pouzin.
The authors wish to thank a number of colleagues for helpful comments during early discussions of international network protocols, especially R. Metcalfe, R. Scantlebury, D. Walden, and H. Zimmerman; D. Davies and L. Pouzin who constructively commented on the fragmentation and accounting issues; and S. Crocker who commented on the creation and destruction of associations.
We began doing concurrent implementations at Stanford, BBN, and University College London. So effort at developing the Internet protocols was international from the beginning.
In the early 1970s Mr Pouzin created an innovative data network that linked locations in France, Italy and Britain. Its simplicity and efficiency pointed the way to a network that could connect not just dozens of machines, but millions of them. It captured the imagination of Dr Cerf and Dr Kahn, who included aspects of its design in the protocols that now power the internet.
The network research community formed the [International] Network Working Group (INWG) ... and out of this came the ... transmission control protocol/internet protocol (TCP/IP)
In chronological order: