A Moment of Collaboration
The overarching theme of this series on the history – and pre-history – of the Internet is conflict: different actors with different goals and visions shaping the medium. The 1960′s saw a tension between the Cold War and the counter-culture. The 1970′s saw the rise of commercial firms since advances in processing power meant that mass-produced “microcomputers” (desktops) could be sold to businesses, rather than a few hugely expensive machines going to universities and the government.
The 1980′s broke this pattern of conflict. It was a rare moment when a single vision – adoption of the values of openness and collaboration for the common good – were unquestionably at the fore. It was in the late 1970′s and early 1980′s that the technical and ideological foundation of the Internet – which began in the high-tech Silicon Valley / high-karma San Francisco milieu of the Bay Area in the 1960′s – were finally “baked in.”
The Internet Learns a Single Free Language
The first critical development was that a software for computer networking and data transfer was developed using government funding and was thus free to use by any hardware or software designer. It was called TCP/IP (Transmission Control Protocol and Internet Protocol) and was extremely robust. In fact, it is extremely robust: Internet Protocol version 4 (IPv4), first defined in 1980, is still running the Internet today. That TCP/IP became the dominant communication protocol of the Internet meant thatthe Internet was born as a public entity rather than a commercial one.
In the late 70′s and early 80′s there were already a small number of computer networks – local area networks (LANs) – that existed in offices or university. The problem was that they all used different languages to talk to one another. Without a common language, it would be impossible to connect the networks to one another and create the “internetwork” that became the Internet.
In 1976 Vint Cerf, now Google’s Chief Internet Evangelist, went to work for DARPA’s Information Processing Techniques Office, the Defense Department project that began the global Internet. DARPA’s vision was still a Cold War one. As Cerf described in 2010, “DARPA was looking for ways to build command-and-control systems that had no central structure, and were highly distributed, and that could be readily reconstituted.” This meant that networks needed to be connected to one another so that there would be multiple paths from any computer to another. Isolated networks had limited value. They were connected, but only within a single institution.
As a result, Cerf and Robert Kahn, co-inventor of TCP/IP and also a scientist at DARPA, began encouraging university LANs to connect to ARPANET, the DARPA computer network that formed the core of the modern Internet. In his book Net Effect, Thomas Streeter explains how Cerf and Kahn used free TCP/IP software – and the desire to connect – as incentives to turn isolated networks into a larger internetworked Internet:
The goal, moreover, was not to be secretive and exclusive. In 1980, when ARPANET’s Vint Cerf met with a group of computer science professors from across the country, he offered to connect the ARPANET to a proposed research network if it adopted TCP/IP protocols. This set the trend towards encouraging open access to the internet, which would become the informal policy throughout the 1980′s….
The Military Gives up Control of the Internet
The first amazing development of the 1980′s was the rise of free software for connecting to and transferring data across the Internet. The second amazing development was that the Internet lost its military character and became a civilian network. In 1983 ARPANET was split into a military and civilian Internet. The civilian Internet moved out of the Department of Defense and into the ownership of the research-focused National Science Foundation. It was the beginning of the Internet that we know today.
Why Did the Government Geeks Beat Out the Capitalists and Cold Warriors?
Why was this able to occur? Why didn’t the military fight for control of the Internet or businesses challenge TCP/IP with proprietary protocols that would have forever fragmented the Internet? One reason is that both the military and commercial firms were distracted. DARPA’s high-profile computing project at the time was Strategic Computing Initiative (SCI), a $1 billion artificial intelligence initiative meant to push back against Japan’s advances in the field.
Commercial firms were still focused on selling individual computers. “The U.S. mainstream,” writes Streeter, “was romancing the entrepreneurial tale of stand-alone microcomputers in the 1980′s.” Since the “Internet” of the time was a research network with few applications composed of unconnected LANs at a few firms and universities and a military-owned ARPANET, one can forgive them for not seeing the commercial value in it.
Firms did build their own proprietary Internet protocol software, like Xerox’s XNS and Apple’s AppleTalk, but this software failed to gain market share because TCP/IP was 1) free 2) worked really well and 3) was first to market because it was the protocol of the ARPANET. Because the deck was already stacked against them, commercial firms were convinced to adopt TCP/IP by the end of the 1980′s.
Despite these contextual factors, it would be unfair to diminish the efforts of individuals to build the open Internet we know today. Most of the men (yes, by far men) who opened the Internet to the public are unknown. Let’s look at some of the key players.
Whereas we now associate the military establishment with secrecy and aggressive isolation, the Pentagon point person for ARPANET was Barry Leiner, a man who, like Kahn and Cerf, convinced many private companies to adopt TCP/IP. His time working in think tanks, private industry and the government made him an excellent bridge-builder and his achievement in encouraging the adoption of TCP/IP was equivalent to “persuading the peoples of the world to stop speaking English, French, Russian or Chinese and instead invent and adopt a new language that works even better and allows everyone to communicate with one another.” It was his close relationship with Steve Wolff of the National Science Foundation that allowed for that smooth transition of the Internet into the civilian sphere. Wolff, who got an electrical engineering degree from Swarthmore, was perfectly positioned to bridge the worlds of academic research and high tech.
The Director of DARPA in the early 1980′s, Robert Cooper, was also more interested in building knowledge than achieving military goals. In a 1993 interview he described his criterion for success of a DARPA project:
There were two criteria… one objective I had in the basic research area was to expand the support mainly to graduate students in computer sciences at the key universities… The other criterion for success… was whether some of these ideas… actually got into military systems, and I don’t think the jury is in on how much of that occurred….
This dedication often took dramatic turns, as this anecdote about Vint Cerf indicates. Cerf is a sharp dresser whose frequent attire is a three-piece suit:
At a 1992 meeting that marked a pivotal juncture for the Internet Protocol, engineers were at one another’s throats over a controversial issue. Dr. Cerf took the podium and… proceeded to strip…. He stopped when he reached a T-shirt emblazoned with “I P on Everything,” an inside joke referring to the ubiquity of the Internet Protocol. The audience, he said, “went nuts,” and the tension dissolved. One member of the audience rushed to the podium and placed a $5 bill in Dr. Cerf’s waistband.
There are a few moments in history where the right mix of contextual factors and individual ability come together to take an action with a dramatic positive impact on global human history. In the history of the Internet, the 1980′s was such a period. It was a moment when the right people were in the right positions and something incredibly unlikely, and incredibly good for humankind, occurred.