Something to Share
When the Defense Department issued a $19,800 contract on December 6, 1967, for the purpose of studying the "design and specification of a computer network," the world didn't take notice. But it should have. For, from that small, four-month study grew the ARPANET. And, from ARPANET emerged the Internet.
Like many information and communications technologies, the Internet we know today grew from seeds planted by the U.S. government. Specifically, the Advanced Research Projects Agency (ARPA) was established in 1957 to respond to the perceived scientific and technological advantage the then-Soviet Union displayed in launching the Sputnik satellite. ARPA was charged to "direct or perform such advanced projects in the field of research and development as the Secretary of Defense shall, from time to time, designate by individual project or by category." In plain language this meant that ARPA, along with the newly created National Aeronautics and Space Administration, was to regain technical superiority for the United States.
Some of those employed by ARPA realized the only way this goal could be achieved was to bring together the brain-power resident in discrete pockets at universities and research institutions spread across the United States. To maximize this sharing of brain-power, it quickly became clear that significant advances in computing technology were required. These
computing advances had to provide avenues for both the sharing of ideas and the sharing of computing power and programs. So, an Information Process Techniques Office (IPTO) was created within ARPA in 1962 to achieve this purpose.
The first head of that office was J.C.R. Licklider. He envisioned an "intergalactic" community which could emerge from a single computer time-sharing system. He thought such a community possible because he held a different view of computers. Instead of thinking of computers as giant calculators, Licklider laid out a vision in which computers would fulfill their greatest promise as a "communication medium between people."
About this same time, a RAND researcher by the name of Paul Baran was working on a classified U.S. Air Force contract, whose purpose was to identify ways to strengthen the Nation's telecommunication infrastructure so that it could survive a nuclear strike. Part of his solution was to develop distributed telecommunication networks. If information was truly distributed, instead of everything flowing into and out of central points, then a network could still work effectively even if some legs of the network were damaged or removed. Implementing such a distributed network involved a technique called packet switching.
While this classified early-1960s work was not directly known to ARPA staff, its ideas of a distributed network using packet switching were key concepts for the ARPANET. Thankfully, for the development of the ARPANET, unclassified work on these concepts was also being done independently by other researchers, such as Donald Davies and Leonard Kleinrock.
Packet switching is an approach that breaks a message down into separate, discrete pieces or packets. Each packet then moves from its point of origin to its destination over any open route, regardless of which path the other packets take. When all the packets arrive at the destination they are reassembled -- and the message is delivered intact.
With these ideas -- packet switching and computers as galactic communication devices -- in place, what was needed were technologies that allowed different computers, from different manufacturers, with different operating systems to communicate with each other. To meet this technological need ARPA decided to "contract out," using a competitive bidding process among 140 potential bidders.
In 1968, $563,000 was committed to a contract with the purpose of designing, constructing, installing, testing, and maintaining four interface message processors (IMPS) that would link computers at the Stanford Research Institute, UC-Santa Barbara, UCLA, and the University of Utah. The contract promised that these IMPS would:
"Provide the communications capability required for the ARPA computer research facilities but...also be a unique prototype of future communications systems."
Thus, from these beginnings the ARPANET was born. On October 15, 1969, on the second try, the IMPS installed at UCLA and the Stanford Research Institute connected and began a communication revolution with the words "log in."