This is the story of the creation of ARPANET, the revolutionary predecessor of the Internet, as told by participants in the events
Arriving at Bolter Hall Institute at the University of California, Los Angeles (UCLA), I climbed the stairs to the third floor in search of room #3420. And then I went into it. From the corridor she didn't seem anything special.
But 50 years ago, on October 29, 1969, something monumental happened. Graduate student Charlie Cline, sitting at an ITT Teletype terminal, made the first digital data transfer for Bill Duvall, a scientist sitting at another computer at Stanford Research Institute (today known as SRI International), in a completely different part of California. This is how the story began , a small network of academic computers that became the forerunner of the Internet.
It cannot be said that at that time this brief act of data transmission thundered throughout the whole world. Even Cline and Duvall couldn't fully appreciate their accomplishment: "I don't remember anything special about that night, and I certainly didn't realize at the time that we had done anything special," Cline says. However, their connection became proof of the feasibility of the concept, which ultimately provided access to almost all the world's information for anyone who owns a computer.
Today, everything from smartphones to automatic garage doors are nodes in a network descended from the one Cline and Duvall were testing that day. And the story of how they determined the first rules for moving bytes around the world is worth listening to - especially when they tell it themselves.
“So that this doesn’t happen again”
And in 1969, many people helped Cline and Duvall make that evening breakthrough on October 29 - including a UCLA professor , with whom, in addition to Cline and Duvall, I spoke at the 50th anniversary. Kleinrock, who still works at the university, said that in a sense, it was a child of the Cold War. When in October 1957 the Soviet blinked in the skies over the United States, shock waves from it passed through both the scientific community and the political establishment.
Room No. 3420, restored in all its splendor from 1969
The launch of Sputnik "found the United States with its pants down, and Eisenhower said, 'Don't let this happen again,'" Kleinrock recalled in our conversation in room 3420, now known as the Internet History Center. Kleinrock. “So in January 1958, he formed the Advanced Research Projects Agency, ARPA, within the Department of Defense to support STEM—the hard sciences studied at U.S. universities and research laboratories.”
By the mid-1960s, ARPA provided funding for the construction of large computers used by researchers at universities and think tanks across the country. ARPA's chief financial officer was Bob Taylor, a key figure in computer history who later ran the PARC laboratory at Xerox. At ARPA, unfortunately, it became clear to him that all these computers spoke different languages and did not know how to communicate with each other.
Taylor hated having to use different terminals to connect to different remote research computers, each running on its own dedicated line. His office was filled with teletype machines.
In 1969, such Teletype terminals were an integral part of computing devices
“I said, man, it’s obvious what needs to be done. Instead of having three terminals, there should be one terminal that goes where you need it,” Taylor told the New York Times in 1999. “This idea is ARPANET.”
Taylor also had more practical reasons for wanting to create a network. He constantly received requests from researchers across the country to fund the purchase of larger and faster . He knew that much of the government-funded computing power was sitting idle, Kleinrock explains. For example, a researcher could be maximizing the capabilities of the computing system at SRIin in California, while at the same time the mainframe at MIT could be sitting idle, say, after hours on the East Coast.
Or it could be that the mainframe contained software in one place that could be useful in other places—like the first ARPA-funded graphics software at the University of Utah. Without such a network, “if I'm at UCLA and I want to do graphics, I'll ask ARPA to buy me the same machine,” Kleinrock says. “Everyone needed everything.” By 1966, ARPA had become tired of such demands.
Leonard Kleinrock
The problem was that all these computers spoke different languages. At the Pentagon, Taylor's computer scientists explained that these research computers all ran different sets of codes. There was no common network language, or protocol, through which computers located far apart could connect and share content or resources.
Soon the situation changed. Taylor persuaded ARPA director Charles Hertzfield to invest a million dollars in developing a new network connecting computers from MIT, UCLA, SRI and elsewhere. Hertzfield obtained the money by taking it from the ballistic missile research program. The Department of Defense justified this cost by the fact that ARPA had the task of creating a “surviving” network that would continue to operate even after one of its parts was destroyed—for example, in a nuclear attack.
ARPA brought in Larry Roberts, an old friend of Kleinrock's from MIT, to manage ARPANET projects. Roberts turned to the works of British computer scientist Donald Davis and American Paul Baran and the data transmission technologies they invented.
And soon Roberts invited Kleinrock to work on the theoretical component of the project. He had been thinking about data transmission over networks since 1962, when he was still at MIT.
“As a graduate student at MIT, I decided to tackle the following problem: I’m surrounded by computers, but they don’t know how to communicate with each other, and I know that sooner or later they will have to,” Kleinrock says. – And no one was engaged in this task. Everyone studied information and coding theory.”
Kleinrock's main contribution to ARPANET was . Back then, the lines were analog and could be leased from AT&T. They worked through switches, meaning a central switch established a dedicated connection between the sender and the recipient, be it two people chatting on the phone or a terminal connecting to a remote mainframe. On these lines, a lot of time was spent in idle time - when no one was speaking words or transmitting bits.
Kleinrock's dissertation at MIT laid down the concepts that would inform the ARPANET project.
Kleinrock considered this a wildly inefficient way to communicate between computers. Queuing theory provided a way to dynamically partition communication lines between data packets from different communication sessions. When one stream of packets is interrupted, another stream can use the same channel. Packets that make up one data session (say, one email) can find their way to the recipient using four different routes. If one route is closed, the network will redirect packets through another.
During our conversation in room 3420, Kleinrock showed me his thesis, bound in red on one of the tables. He published his research in book form in 1964.
In such a new type of network, data movement was directed not by a central switch, but by devices located at network nodes. In 1969 these devices were called , “interface message handlers”. Each such machine was a modified, heavy-duty version of the Honeywell DDP-516 computer, which contained special equipment for network management.
Kleinrock delivered the first IMP to UCLA on the first Monday in September in 1969. Today it stands monolithically in the corner of room 3420 in Bolter Hall, where it has been restored to its original appearance, as it was when processing the first Internet transmissions 50 years ago.
"15-hour workdays, every day"
In the fall of 1969, Charlie Cline was a graduate student trying to earn an engineering degree. His group was transferred to the ARPANET project after Kleinrock received government funding to develop the network. In August, Kline and others were actively working on preparing software for the Sigma 7 mainframe to interface with IMP. Since there was no standard communication interface between computers and IMPs—Bob Metcalfe and David Boggs would not invent Ethernet until 1973—the team created a 5-meter cable from scratch to communicate between the computers. Now they only needed another computer to exchange information.
Charlie Cline
The second research center to receive an IMP was SRI (this happened in early October). For Bill Duvall, the event marked the start of preparations for the first data transfer from UCLA to SRI, on their SDS 940. Teams at both institutions, he said, were working hard to achieve the first successful data transfer by October 21.
“I went into the project, developed and implemented the required software, and it was the kind of process that sometimes happens in software development - 15-hour days, every day, until you're done,” he recalls.
As Halloween approaches, the pace of development at both institutions accelerates. And the teams were ready even before the deadline.
“Now we had two nodes, we leased the line from AT&T, and we were expecting amazing speeds of 50 bits per second,” Kleinrock says. “And we were ready to do it, to log in.”
“We scheduled the first test for October 29,” adds Duval. – At that time it was pre-alpha. And we thought, okay, we have three test days to get it all up and running.”
On the evening of the 29th, Kline worked late - as did Duvall at SRI. They planned to try to transmit the first message over the ARPANET in the evening, so as not to ruin anyone's work if the computer suddenly “crash”. In room 3420, Cline sat alone in front of an ITT Teletype terminal connected to a computer.
And here's what happened that evening - including one of the historic computer failures in computing history - in the words of Kline and Duvall themselves:
Kline: I logged into Sigma 7 OS and then ran a program I had written that allowed me to command a test packet to be sent to SRI. Meanwhile, Bill Duvall at SRI started a program that accepted incoming connections. And we talked on the phone at the same time.
We had a few problems at first. We had a problem with code translation because our system used (extended BCD), a standard used by IBM and Sigma 7. But the computer in SRI used (Standard American Code for Information Interchange), which later became the standard for the ARPANET, and then the whole world.
Having dealt with several of these problems, we tried to log in. And to do this you had to type the word “login”. The system at SRI was programmed to intelligently recognize available commands. In advanced mode, when you first typed L, then O, then G, she understood that you probably meant LOGIN, and she herself added IN. So I entered L.
I was on the line with Duvall from SRI, and I said, “Did you get the L?” He says, “Yeah.” I said that I saw the L come back and print out on my terminal. And I pressed O and it said, "'O' came." And I pressed G, and he said, “Wait a minute, my system has crashed here.”
Bill Duvall
After a couple of letters, a buffer overflow occurred. It was very easy to find and fix, and basically everything was back up and running after that. I mention this because that's not what this whole story is about. The story of how ARPANET works.
Kline: He had a small error, and he dealt with it in about 20 minutes, and tried to start everything again. He needed to tweak the software. I needed to check my software again. He called me back and we tried again. We started again, I typed L, O, G and this time I got the answer "IN".
"Just engineers at work"
The first connection took place at half past ten in the evening Pacific time. Kline was then able to log into the SRI computer account that Duvall had created for him and run programs using the system resources of a computer located 560 km up the coast from UCLA. A small part of ARPANET's mission was accomplished.
“By then it was late, so I went home,” Kline told me.
The sign in room 3420 explains what happened here
The team knew they had achieved success, but did not think much about the scale of the achievement. “It was just engineers at work,” Kleinrock said. Duvall saw October 29 as simply one step in a larger, more complex task of linking computers together into a network. Kleinrock's work focused on how to route data packets across networks, while SRI researchers worked on what makes up a packet and how the data within it is organized.
“Basically, that's where the paradigm that we see on the Internet was first created, with links to documents and all that stuff,” Duvall says. “We always imagined several workstations and people interconnected. Back then we called them knowledge centers because our orientation was academic.”
Within weeks of the first successful exchange of data between Cline and Duvall, the ARPA network expanded to include computers from the University of California, Santa Barbara, and the University of Utah. ARPANET then expanded further into the 70s and much of the 1980s, linking more and more government and academic computers together. And then the concepts developed in ARPANET will be applied to the Internet that we know today.
In 1969, a UCLA press release touted the new ARPANET. “Computer networks are still in their infancy,” Kleinrock wrote at the time. “But as they grow in size and complexity, we are likely to see the proliferation of 'computer services' that, much like today's electrical and telephone services, will serve individual homes and offices across the country.”
Today this concept seems quite old-fashioned - data networks have penetrated not only into homes and offices, but also into the smallest devices belonging to the Internet of Things. However, Kleinrock's statement about "computer services" was surprisingly prescient, given that the modern commercial Internet did not emerge until several decades later. This idea remains relevant in 2019, when computing resources are approaching the same ubiquitous, taken-for-granted state as electricity.
Perhaps anniversaries like this are a good opportunity not only to remember how we came to this highly connected era, but also to look to the future - as Kleinrock did - to think about where the network might go next.
Source: habr.com