How did computers originate?

In 1942, Licklider received a PhD in behavioral psychology from the University of Rochester. He first worked as a research assistant at Swarthmore College and later at Harvard University. Became a researcher at the Psycho-acoustic Laboratory. There, he served as a lecturer until 1951. Later, he went to the Massachusetts Institute of Technology, where he conducted research on "listening" and "speaking."

His office is in the basement of Lincoln Laboratory. At that time, all the rooms in the basement had their doors open except one. A young computer expert, Wesley Clark, finally decided to open the door and take a look after many days of hesitation. As a result, Licklider was conducting psychological testing experiments inside. Clark told Licklider that the same result could be obtained using his TX-2 computer.

Suddenly, they became good friends, and Licklider's interest also turned to computers. This TX-2 computer had 64K of memory, equivalent to the small calculators we carry in our pockets today. But at that time, computer operators could only walk through its "belly" - because this computer took up two entire rooms! Even for someone with a high IQ like Lick Ridder, operating a computer is not an easy task. There is much to learn.

It seems that Licklider is a very far-sighted person. Many developed countries continued to conduct research around analog equipment until the early 1990s. Licklider's interest shifted from analog equipment to digital equipment starting in 1957. During this period, he joined BBN Company (BoltBeranek and Newman, Inc.), which later designed and manufactured the earliest computers for the Internet.

Even the process of Licklider's decision to take up a position at DARPA is quite legendary. Katie Hafner and Matthew Lyon introduced the book "Where Wizards Stay Up Late—The Origins of the Internet" (Simon & Schuster, 1996). The situation at that time.

In 1962, DARPA’s third director, Jack Ruina, called on Lick Riddell, who was working at BBN, and his friend and colleague Fred Frick, who was working at Lincoln Laboratory, to discuss building a Department to study "command and control" technology.

Likridel originally just wanted to listen. But I was quickly attracted by this question. In his view, the problem of "command and control" is also the problem of "Human-Computer Interaction". And that was the question that interested him.

However, it is one thing to be interested, and another thing to engage in work in this field. Both Licklider and Frick are very busy and have their own jobs that they cannot get away from. And "command and control research" was so important, at Ruina's insistence, the two had to toss a coin to decide who would put down the work at hand to lead this department.

Speaking of which, Licklider "deserved" to go to the Ministry of Defense. After his new job was "decided" by the coin, Licklider put forward two conditions: first, he would only work at DARPA for two years and then return to BBN. Second, he needs to be able to handle the affairs of this department with full authority, and no one else can interfere.

Later facts proved that DARPA had found the right person in finding him to lead this key department.

Licklider is an easygoing person. Anyone who met him for the first time was told not to call him by his full name, but to call him "Lick."

Many people were deeply impressed by his easy-going personality.

As an official of the Ministry of Defense, Licklider has extensive connections with the military, which allows him to obtain large sums of funds for academic research from the military budget; and as a scholar, he Inseparable from the academic world, his academic background made it possible for him to bring an air of academic freedom on campus to the highly disciplined U.S. military. It was Licklider's efforts that directly promoted DARPA's long-lasting and effective large investments in the field of information technology. It is precisely because of his influence that a group of elites can gather under the banner of DARPA; it is also because of his influence that DARPA can trust these elites and not set specific research goals for them, allowing geniuses to have Possibility of free play.

Alan Perlis, a researcher at the time, later recalled:

“I think we should all be very satisfied with ARPA. Because ARPA did not specifically ask us to do this or that. -- Let's do workstations. There's never been an order like, 'We need a plan for workstations.' God knows, if they had asked for that, they would have had a bunch of plans for workstations. It was because of Licklider that ARPA understood that as long as some outstanding people work together to study computers, excellent results will be achieved. "So, "We all owe ARPA a lot, because they do not impose rigid requirements on our tasks. I want to believe that the purpose of the military is to support ARPA, and the purpose of ARPA is to support academics." (munity) This is what distinguishes him from ordinary computer experts. Licklider and Robert Taylor also wrote a special article "The Computer as a Communication Device" to discuss the role of computers in human communication. These ideas of his undoubtedly played an important role in establishing DARPA's original guiding ideology.

The difference between the "Information Processing Technology Office" and the general computer research department can also be proved from another perspective. In 1963, when the Information Processing Technology Office was first established, the head of DARPA had questions about the role of this department. In their view, if the computer industry can do it, there is no need for DARPA to do it. "If this thing is worth doing, the computer industry will do it. Then there is no need for us to support such a thing." (DARPA original material, III-23)

They obviously did not understand , the work of the "Information Processing Technology Office" was not what the computer industry thought of doing from the beginning. Because this office is not just about studying computer technology issues, but about making computers a tool for human communication. As written in the "Report to End ARPANET":

"ARPA's goal is to make computers an intermediary for people to communicate," and "the computer industry still regards computers mainly as computing tools. This The stereotype shows up even in the communications systems they recently designed. "…"Even in universities, or at least in some universities, many people still insist on the concept of computers as computing tools." (DARPA. Original data, III-24)

The so-called "communication" cannot of course be the communication of a computer. If you want to communicate, you have to build a network.

In September 1964, the Second Information Systems Science Conference was held in Virginia. During the meeting, Larry Roberts had informal conversations with Lick Ridder, Fernando Corbato and Alan Perlis, and confirmed such a basic principle:

“The most important problem we currently face in the computer field is the network. , which means that you can easily and economically connect from one computer to another to achieve maximum sharing of resources.

"(book/ch.4_Arpa2Usenet.html)

The glorious mission of realizing this ideal has historically fallen on the shoulders of the Advanced Research Projects Agency and the Information Processing Technology Office (IPTO) of the U.S. Department of Defense. At that time, IPTO directors during the establishment of the network for DARPA were: Lick Ridder (1962-1964), Ivan Sutherland (1964-1966), Robert Taylor (1966-1969), and Lawrence Roberts (1969-1973). From 1974 to 1976, Licklider made another comeback, and this time he was replaced by C. Russell (1976-1979).

1966 was an important year for DARPA. At the beginning of the year, Robert Taylor became the director of IPTO.

The director of DARPA was also replaced by Charles Herzfeld, a physicist from Austria. There is a joke that if you are interested in research. If you have a good idea for the plan, just go to Herzfeld and you can get the money in less than 30 minutes!

The book "The Place to Contain the Magic - The Origin of the Internet" introduces DARPA's first effort in building a network! How did the money come about? One day in 1966, Robert Taylor went to see Herzfeld. The problem was obvious: people working with IPTO were increasingly demanding more computers. Besides, everyone couldn't afford it. They also need to understand each other's work, and it is best to cooperate with each other.

Herzfeld asked: "Is this difficult? "

Answer: "Oh, it's not difficult. We already know what to do.

"Great idea! Go ahead and do it. Now you've added another $1 million to your budget. Go for it!"

When Robert Taylor left Herzfeld's office When I came out, I felt a little regretful and said to myself: "We only talked for less than 20 minutes!"

However, just having money is not enough, you need to find someone who can fully An outstanding and far-sighted computer engineer who understands Licklider's idea of ??building a network and can implement it to the end

Three visits to the thatched cottage

Although Taylor has already made plans in his heart, Larry Roberts was the best person to design the network for DARPA; however, later facts proved that it was more difficult to hire Roberts to work for DARPA than it was when Liu Bei hired Zhuge Liang to "visit the thatched cottage three times".

Larry Roberts is the son of a chemist at Yale University. First I went to MIT and learned to play with the TX-0 computer there. Later, he went to Lincoln Laboratory and programmed a complete operating system program for the TX-2, the most advanced computer at the time.