Daniel Belin's answer to Who invented the internet? has most of this right. Bolt, Beranek & Newman is the correct spelling for BBN. ARPANET was a homogeneous network. Steve Crocker led the development of the host protocols for ARPANET and formed the Network Working Group made up of many from the institutions funded by Defense Advanced Research Projects Agency (DARPA) to do research in Computer Science and Artificial Intelligence. Bob Kahn was a key ARPANET architect at BBN but joined ARPA in late 1972. He initiated the INTERNETTING research program and I received funding at Stanford University to work on the design of a system of heterogeneous networks. Bob and I wrote the first description of the Transmission Control Protocol (TCP) in September 1973 and reviewed it with the International Network Working Group that I chaired and which was set up at the time of the first public demonstration of the ARPANET in October 1972 (Bob Kahn organized this with Larry Roberts' encouragement). In May, 1974, the more formal paper on TCP was published in the IEEE Transactions on Communications.
There were many people who were involved in refining the TCP and splitting it into TCP/IP. The Internet Society has recognized many of them in its Internet Hall of Fame as pioneers. Jon Postel, Dave Clark, Bob Braden, Danny Cohen, David Reed, Noel Chiappa, and many others. There is a plaque at Stanford that references many who worked specifically on TCP in the first three years or so of the project from 1973 to 1976.
Also, I can confirm that Bob Kahn and I wrote a response to the unfair criticism of Al Gore which Kron has correctly identified. You can find a copy of that response athttp://amsterdam.nettime.org/Lis....
using the Internet since it was called ARPAnet
The Internet was created as a combined effort between private industry, including companies like Digital Equipment Corporation, and the Defense Advanced Research Projects Agency, an arm of the US government that does military research.
It began as a network to unite mainframe computers at schools such as MIT and Cal Tech and government installations, so that files could be shared and information copied. That first early network was called ARPAnet.
The backbone of that early network ran largely on DECsystem-20 mainframes, the first mainframe computer I learned to program.
It grew and grew, eventually linking personal computers, first through dialup modems and then through broadband.
The actual concept for a global packet-switched network, which has now incarnated itself as The Internet, originated with an Massachusetts Institute of Technology computer scientist and well known computer luminary, J.C.R Licklider (commonly known as "The Lick"). He published the original proposals for something known as the Intergalactic Computer Network in 1962 in a series of memos from the company Bolt, Beranek & Newman (BBN). In these ideas, he basically predicted the modern Internet as we know it today. (Fun fact: he also innovated national public broadcasting as we know it.)
Shortly thereafter in 1967, the Advanced Research Projects Agency's (now DARPA) Information Processing Techniques Office (IPTO, now I2O) sent out several Requests for Quotes to institutions to write the initial specifications for the routing machines. The contract was won by the Stanford Research Institute (now SRI), and they produced the initial specifications for the Interface Message Processor (IMP). Shortly thereafter, another contract was won by Lick's former company Bolt, Beranek & Newman to construct it. This would begin the foundations for the modern Internet.
The first two Interface Message Processors (IMPs) were installed at SRI and University of California, Los Angeles (UCLA). The IMP at SRI was connected to an SDS 940, while the UCLA IMP was connected to a SIGMA 7. The UCLA/ARPA team was the stomping grounds for many of the founding members of the Internet, as Dr. Len Kleinrock's team included Vint Cerf, Jon Postel, and many others. This membership on the team would be Vinton Cerf's first insight into ARPANET. Incidentally, one of Vint Cerf's responsibilities would be to write the Sigma 7 EXperimental Timesharing System's manual, a.k.a., The SEX Manual. This document would be one of the few to share authorship of almost all the fathers of the Internet, including: Vint Cerf, Jon Postel, Charlie Kline, as well as Steve and Dave Crocker.
The first message on the ARPAnet was the sending of the letters L and O (of the full word login) before the system crashed. These words were sent between UCLA and SRI on October 29, 1969 at 22:30. An hour later, the full message "LOGIN" was sent after the SDS computer recovered from its error. This is the true "Hello Dolly" moment of the Internet. The permanent link between UCLA and SRI was established about a month later, much to the chagrin of AT&T (company) Long Lines, on November 21st.
This permanent link began a large period of growth for the ARPAnet. Before the year was out, a four-node network was connecting SRI and UCLA to University of California, Santa Barbara (UCBS) and the The University of Utah. The fifth link was the first intercontinental link between BBN and UCLA. The network kept growing, with 22 nodes in 1972 and over 57 IMPs in 1975. Soon after, it was declared operational and functional ownership of the ARPAnet was transferred from DARPA to the Defense Communications Agency.
In 1973, Bob Kahn and Vint Cerf joined forces to bring the world TCP/IP. TCP/IP was conceptualized to replace the inadequate Network Control Program (NCP). The key transition in TCP/IP was that the network would no longer be responsible for reliability, instead, the end hosts would be responsible for reliability. In addition, it would hide differences at lower levels of the network by providing a common protocol which could then be encapsulated and transmitted at lower levels. These two factors were key to the success of TCP.
Over the course of the next seven years, they developed TCP/IP. The first major breakthrough for TCP/IP came in 1982 when the U.S. Department of Defense declared TCP/IP the standard for Defense computer networking over competing protocol stacks such as OSI. This would soon lead to what many consider the birth of the worldwide TCP/IP Internet.
On New Years Day, 1983, the Internet had arrived. NCP ceased to function on the ARPAnet and TCP/IP took over. This was a major flag day for the ARPAnet, as it adopted Cerf's and Kahn's internetworking protocol stack. While this event was the epoch for the formation of the Internet, it would be the beginning of the sunset for ARPAnet. In 1981, the formation of CSNET meant that there was now a competing academic network to ARPAnet. In addition, the DoD was beginning to have concerns about hosting an academic and research network. In 1985, NSFNET was formed, which was to become the backbone of what we know as the Internet today. This, combined with the splitting of MILNET from ARPAnet and the formation of the Defense Data Network led to the slow shutdown of the ARPAnet.
With research institutions connecting mainly to NSFnet and other international research networks, and the DoD using MILNET and the Defense Data Network, ARPAnet was mostly defunct by 1986. The final IMPs were shut down in 1989, and the official deactivation of ARPAnet occurred on February 28th, 1990. Vint Cerf commemorated the deactivation in the "Requiem of the ARPAnet":
It was the first, and being first, was best, but now we lay it down to ever rest. Now pause with me a moment, shed some tears. For auld lang syne, for love, for years and years of faithful service, duty done, I weep. Lay down thy packet, now, O friend, and sleep -Vinton G. Cerf
After this, the Internet blossomed (with the help of Al Gore). International links were established; the internet was made public; and through Mr. Gore's legislation in the early 1990s, commerce on the internet began. NSFNet spun into an academic backbone which began to connect into first the major telcos, then an ecosystem of hundreds of service providers and technology companies.
The rest, they say, is history!