50 years ago there was no expectation of where the achievement would lead, and the real impact would not begin to be understood for almost 20 years.
By Harlan Lebo
Cultural Historian and Author
October 29, 1969
Boelter Hall is a nondescript but pleasant enough brick building on the UCLA campus, framed by California olive trees and bordered by the grass-lined walkway known as the Court of Sciences in the south section of the university.
During the day, Boelter Hall teems with engineering students. But in the evening, with most student housing far across the campus, the building descends into quiet solitude – an ideal setting for work that requires time and focus.
October 29 was a perfect night to change the world.
Charley Kline, a graduate student in the engineering school’s computer science department, viewed the late hours as a time to work in Boelter free of distractions.
“I was a tech guy who liked to program at all hours,” Kline said, “and it was much easier for me to stay focused in the middle of the night.”
But “to program” in 1969 was vastly different from today’s UCLA engineering student, pecking away on a two-pound laptop while sitting in a coffee shop in Westwood a few blocks from campus. For Kline, and a generation of students studying in the young field of computer science, a PC of any size or price was almost a decade away; “to program” meant working in an on-campus laboratory, “computers” were room-filling systems, “keyboards” usually meant cumbersome stand-alone terminals shrouded in sheet steel.
That night, Kline’s project would be relatively simple, and involved testing a new system that had been installed at UCLA for almost two months: an experimental project funded by the federal government to create links between computers in locations across the country.
Simple – if it worked.
For Kline, such assignments were departures from the traditional education that most computer science students pursued in the 1960s. Although some opportunities in computing involved government or academic systems that were used for scientific research and calculation, in that era, jobs in computer science generally meant support for large systems that served as giant calculators and billing machines for banking and other industries.
But Kline sought different types of opportunities.
“I was interested in exploring the problems that were emerging in a world where computers worked independently with some success, but had a great deal of trouble communicating with each other,” Kline recalled. “Our goal was to determine how to make them talk.”
Kline found an opportunity for that mission in the laboratory of Leonard Kleinrock, who at 35 was already recognized for developing a mathematical theory of the methods to create communication pathways between computers at different locations.
In 1969, Kleinrock’s principal project to validate his theoretical discoveries was to participate in building an experimental network, a system that would, according to the July 3 press release, “for the first time, link together computers of different makes and using different machine languages.”
“As of now, computer networks are still in their infancy,” Kleinrock explained in the release, “but as they grow up and become more sophisticated, we will probably see the spread of ‘computer utilities,’ which like present electric and telephone utilities, will service individual homes and offices across the country.”
Creation of the network, reported the release, “represents a major forward step in computer technology and may serve as the forerunner of large computer networks of the future.”
Almost 50 years later, Kleinrock recalled, “In simplest terms, we were trying to shift the thinking from everyone using a large stand-alone computer, to a linked network that could exchange information.”
But for the moment, functional networks were still to come; first came learning how to create practical connections between computers, with a goal of linking systems at universities, government agencies, and scientific institutions so they could communicate and exchange information. To start, four computers would serve as the foundation of the system: machines at UCLA, the Stanford Research Institute, UC Santa Barbara, and the University of Utah.
* * * * * * * *
At UCLA, just receiving the delivery of the computer in 1967 became a logistical headache; the computer – a Sigma 7 built by Scientific Data Systems of Santa Monica – was eight feet wide and almost six feet tall, so cumbersome that no elevator in Boelter Hall could accommodate it. To move the Sigma 7 into the building required a forklift on the loading dock at the back side of the building to raise the plastic-wrapped computer to the third floor, where a section of railing had been ripped out to allow the equipment to slide through.
In the lab, the Sigma 7 was connected to a Honeywell DDP-516, a “mini-computer” (merely the size of a refrigerator). The Honeywell was chosen not only for its price and performance, but also for its rugged structure built to military specifications; to demonstrate to visitors the computer’s physical strength, Kleinrock would pound on the cabinet with his fist.
The Honeywell was equipped with additional technology created by the consulting firm of Bolt, Beranek, and Newman, a Cambridge, Massachusetts company that added the parts to transform the computer into an “Interface Message Processor,” better known as an IMP (this was the first device now called a “router”).
When attached to the Sigma 7, the IMP would – everyone hoped – serve as an all-purpose gateway that would link computers in many locations, built by separate manufacturers, created for a range of purposes, and all using different types of programming languages. It would be an ambitious project.
The first IMP – today still identified with the tag that marked it as node #1 in the national network to come – had been delivered to UCLA on August 30; three days, later, Kleinrock’s team successfully linked the IMP to the Sigma 7. With each IMP requiring a month to construct, the second was ready late in September. On October 1, it was delivered to the Stanford Research Institute in Menlo Park, 350 miles north of UCLA. Later, the third and fourth IMPs would be sent to UC Santa Barbara and the University of Utah, completing the equipment for the quartet of computers that would be the start of the new network.
The next step was to encourage the machines to talk, listen, and respond.
Around 9 pm, Kline walked across the Court of Sciences to the entrance of Boelter, and then downstairs to room 3420, the home to the UCLA Network Measurement Laboratory, where the new computers had been shoehorned through the door.
Kline sat down at the industrial-metal desk next to his terminal, picked up the phone, and dialed a number in Menlo Park.
* * * * * * * *
At the Stanford Research Institute, Bill Duvall was waiting for Kline’s call. Duvall, at 25, was working full-time at the institute. A nonprofit research organization in Menlo Park that was spun off from Stanford University in 1946, the Institute (now known as SRI), had been established to serve as a “center of innovation” – an organization-for-hire that conducted research, designed products, and developed plans for civic agencies and private industry.
For Kline and Duvall, the task that night was clear.
“Our goal was to test the capability of the UCLA machine to log in to the computer at SRI,” said Kline.
It would have seemed a simple experiment, but in practice, the process was much more complicated. That night they would try it.
* * * * * * * *
At 9:30 p.m., Kline and Duvall, each on a telephone headset, powered up their equipment and activated their experimental operating systems that would allow Kline to connect.
Just after 9:30, Kline typed a letter.
“The first letter I typed was an L,” Kline said. On Duvall’s terminal, the “L” appeared.
Kline tried again; he typed the “O.”
“I got the ‘O,’” Duvall reported.
But that was all; the computer at SRI overloaded and the connection crashed. Two letters was as far as they got.
But two letters were enough. For at least a moment, the connection had worked. The first communication between the computers was “LO” – an inadvertent, almost-biblical declaration of the beginning of a new age.
“We couldn’t have planned a more powerful, more succinct, more prophetic message,” Kleinrock remembered.
Duvall was able to quickly fix the problem, and an hour later the two machines were again connected, with Kline successfully logging in. At 10:30 p.m., Kline duly recorded the moment by writing the result in the laboratory’s logbook (see image at top of article), and then went home to bed.
Duvall did not see the need for festivities either. He stopped by a local hangout for a burger and a beer.
“It was no celebration,” Duvall said. “I was hungry.”
* * * * * * * *
Kline and Duvall did not mark the moment because they did not realize they had a reason to celebrate. The pair viewed their work that night as simply another step in what they knew would become a long and complex series of technological events.
But leading to what? The link between computers at UCLA and SRI was never intended to become the indispensable technology used daily by billions. Kline, Duvall, Kleinrock, and hundreds of other computer scientists developing 1960s technology had hopes of building a system that would, perhaps at most, connect computers so they could easily exchange information and allow their users to communicate with each other.
At a time when the first personal computers would not appear until the late 1970s, and public access to the internet would not be available for almost 15 years after that, a future filled with billions of websites, online shopping, social media, and instant global access to information was not even the remotest practical consideration. Such miracles were being pondered only as fanciful theory – and on a much more limited scale – by a handful of visionaries.
The birth of the internet – if one technological link in a long chain can be described as a “birth” – had no emotion associated with it; there were no ticker-tape parades, no drama of Thomas Edison watching the first light bulb burn while he contemplated the enormity of what he had done. But the connection achieved on October 29, 1969 was, if nothing else, the starting point of a journey leading to technology that not only succeeded in its original objective, but would evolve into a phenomenon for communication beyond anyone’s most outrageous expectations.
Like all great technological achievements, the internet as we know it exists thanks to a serendipitous intersection of events, people, and inspiration. Over the decades since, some of those combinations would thrive and change the most fundamental activities of work, play, and human interaction; others would fail spectacularly.
Perhaps the most astonishing issue yet to come about the internet would be the extraordinary story of its emergence, as it progressed from a fragile connection between two computers in 1969 with a modest intended function into the most pervasive communications tool of its age – possibly of any age – affecting everything we do, everything we say, and everything we achieve.
The internet serves as an instrument for soaring to creative heights, and spotlights troubling questions about the lowest forms of human depravity. It produces unprecedented opportunities for social interaction while raising deep questions about personal privacy and national security. And because of the internet, perhaps more than any other human advancement, the world is now a much different place than before it arrived, and continues to be reshaped as the technology evolves.
But on October 29, 1969, all of that was years in the future. If there was a single moment that would define the start of the technology that would become the internet – this was it: the future was born.