text-only page produced automatically by LIFT Text
Transcoder Skip all navigation and go to page contentSkip top navigation and go to directorate navigationSkip top navigation and go to page navigation
National Science Foundation
Special Report
design element
NSF and the Birth of the Internet — Home

NSF and the Birth of the Internet Text-only | Flash Special Report

Computing Grows in Importance
The 1970s saw computers grow in strength, speed, and capabilities. They also began to be integrated into everyday life. By the end of the decade, researchers at large companies like IBM and Xerox as well as hobbyists and backyard inventors like Steve Jobs and Bill Gates were building smaller but still powerful computers and software that businesses and individuals could use. Though not quite personal, computers were changing the way people worked and did business, and the number of people who used computers began to grow.

Image: Office workers check the printer output on an IBM Model 138 computer. Introduced in 1976, the Model 138 was similar to other computers introduced in the mid-1970s—faster, more powerful, and easier to use than previous computers. As a result, more and more companies began using computers to speed up their operations and improve their efficiency. 
Credit: IBM Archive

ARPANET Expands and Advances
Meanwhile, ARPANET began to grow rapidly as more universities and government research laboratories were hooked up to the network using tools such as TCP/IP and other innovations that allowed computers to send and receive data to other computers across the county.

Image: A computer operator at the National Center for Atmospheric Research (NCAR) makes adjustments to a CRAY-1 supercomputer in the mid to late 1970s. The CRAY-1 machine provided NCAR with invaluable computing for its early efforts in tropical numerical weather prediction. Researchers across the country were beginning to see the potential of using computer networks to access these large and powerful machines from remote locations.
Credit: © University Corporation for Atmospheric Research

First Email
In 1972, researchers sent out the first email message using the now-familiar name@domain.com nomenclature we still use today. By the end of the decade, thousands of individuals across the country will be able to send and receive email. Most of these users are scientists and engineers, who use the speed of email to quickly collaborate with colleagues around the country.

Image: This photo shows the two computers that sent and received the first email message using the ARPANET. The computer in the background, the BBN-TENEXB, sent the message received by the BBN-TENEXA computer in the foreground. The teletype machine on the left was a Teletype KSR-33 terminal that printed the first email.
Credit: ©1971 Dan Murphy

Network Envy
Even as ARPANET grew in size and complexity, it was only open to select organizations and institutions. This excluded a large number of computer scientists and computing centers that wanted to reap the benefits of such a robust network. By the middle of the decade, ARPA, believed that its network should be maintained by another entity.

Image: A computer-generated global weather model created by Warren Washington in the 1970s. Around the world, computer scientists and researchers saw the power and potential that computer networking had to speed collaboration and share powerful computing resources.
Credit: © University Corporation for Atmospheric Research

Other Networks Arise
The 1970s showed the value and feasibility of large computer networks, but it the networks were still relatively expensive to create and maintain. By the end of the 1970s, NSF began looking into supporting other research networks that would allow more universities and research centers to be networked together. At the same time, computer networks began to spring up in Europe and Asia.

Image: An illustration of the Global Atmospheric Research Program (GARP). Developed in the 1970s by the Global Atmospheric Research Program, GARP used a network of earth-bound sensors, satellites, and research centers to study global atmospheric circulation. It was one example of how researchers were using the power and promise of networking to approach complex scientific concepts and challenges.
Credit: © University Corporation for Atmospheric Research

Campus Computer Networks
During this time, several universities began building their own computer networks to aid research. George Strawn, currently the Chief Information Officer at NSF, was a computer scientist at the University of Iowa in the 1970s. He and colleagues at other universities saw the potential that computer networks had to offer as they began to use email and other innovations. At the same time, the development of micro computers meant that universities were moving away from large research computers in favor of many smaller computers dispersed around campus. This posed a problem for researchers who still needed access to faster, stronger computers and it opened an opportunity to expand fledging networks to more and more users.

Video Transcript

George Strawn, National Science Foundation:  In, uh, 1970, I became an Assistant Professor of Computer Science and an Assistant Director of the Academic Computation Center at Iowa State, and computer networking was on our screen right from, uh, the early 70s.  Computers in those days were a little different than now, as I’m sure you know.  We had one academic computer on campus, which operated at about a million instructions per second and cost several millions of dollars.  Compared to today, a desktop computer or a laptop operates at a billion instructions per second and costs less than $1,000.00 so there’s been a factor of a million improvement in price performance of these things.  Especially since computers were relatively new in 1970, university administrators were saying, “Goodness, these things are expensive.  Can’t you share?”  So, uh, the administrators wanted Iowa and Iowa State to share a computer.  Well, we knew that we really couldn’t do so but we wanted to show cooperation and ability to use our resources to the greatest.  So, we hit upon a scheme of connecting the two computers, a two-node network, if you will, so that, uh, so that we could load balance and if one computer was unloaded and the other one was more loaded, we would ship jobs one way or the other.  This would be what the Internet would later call file transfer.  Especially in the early 70s, and in most places, uh, most large universities, um, the, um, mainframes were what we had and even minicomputers didn’t come in at a place like Iowa State until the late 70s.  So, here was this one large computer.  It had punch cards in and line printer output out.  The first step that we made toward using computer networks to help the process was to put remote job entry stations around campus.  Before we were done, we had about ten sites in ten buildings, so you had to walk much less and if you were lucky and in one of those buildings that was heavy users, you would, uh, not even have to leave the building in inclimate weather and you could have remote card input and remote printer output.  Uh, we would consider that pretty primitive use of computer networking but that’s the way it began on campus for campus-wide network – remote card readers and remote line printers.  Submitting computer programs and getting the output back was the primary job of computing in those days and, um, it was, first step was to have the remote job entry but it would’ve been so much better, and everybody knew it would be so much better, if you could interact directly with the computer rather than through cards.  So, the first step would be to replace card input and printer output with terminal input and terminal output.  First, sort of Teletype type devices and later CRT devices.  That is, that’s similar to the screen found on a PC of today.  Since you were at the terminal too hard to have a email scheme, uh, introduced where if you had an account – everybody had an account name and account number so you could, as long as you knew the account names of your correspondents, send them email message, very much the way we send email today.  So, uh, I began in the mid-70s communicating with my students in my computer science classes with email, and, uh, that was a good adjunct to, uh, to office hours.  People could – could write in questions and I could write back answers and so on and so forth, very much like, uh, like email is done today, but – but quite a while before email came on the consciousness of the public.  I do remember that as with the, um, Aripnet, which was experiencing email at just about the same time in the mid-90s, email was viewed as a, um, somewhat of an interruption of real work and you wanted to make sure that email didn’t clog up the system from getting its real work done, and – and there was even sometimes moves to, uh, to suppress the use of email.

Video Credit: Cliff Braverman, Dena Headlee, Lauren Kitchen and Dana Cruikshank for National Science Foundation
John Prusinski, Kathryn Sharar Prusinski, Michael Conlon and Joan Endres for S2N Media

Still Image Credit: © 2007 Jupiter Images Corporation