Issue No. 06 - June (2006 vol. 39)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/MC.2006.210
David Alan Grier , George Washington University
For a year, I worked in an office with a view across a four-foot-wide airshaft into an apartment that was shared by two young men. Neither of these men was older than 25, and both had apparently taken a vow to live public lives. They never closed the curtains on their windows, if indeed they had curtains to close, which gave me a detailed view of their home. The decorations showed the tasteful touch of the male hand. The walls were covered with posters that proclaimed the pair's interests in medievalist fantasy drama, women's specialty fashion, and sports sponsored by brewed malt beverage companies.
A second wall held a large video game console, which commanded much of their attention. They spent a great deal of time in front of the screen, sitting on an old couch and playing round after round of games. Their bodies swayed in time to the simulated motion of the games, and their excited voices would occasionally penetrate the windows of their apartment.
Many a day, they would start in the late afternoon and continue their play into the night. More than once, I opened my office in the morning and spied one of the two, sitting on the couch with a game controller clenched in his hands as he struggled to stay awake while mastering some aspect of a game. Observing this scene, I could only wonder how these video games had so completely captured his imagination.
Although computer games are now a form of entertainment, they actually grew out of a series of efforts to develop computer-based methods to represent the natural world, abstract complex activity, and analyze organizational problems.
The game of chess quickly captured the attention of early computer researchers. "Although perhaps of no practical importance," wrote Claude Shannon of Bell Telephone Laboratories, "the question is of theoretical interest and it is hoped that a satisfactory solution of this problem will act as a wedge in attacking other problems of a similar nature and of greater significance."
Whether chess-playing computers were of practical importance or not, both Shannon and Alan Turing, the English mathematician, studied them. Between 1946 and 1954, these two researchers developed the fundamental theory of chess-playing programs. In 1950, Shannon outlined his ideas in a published paper, and Turing incorporated his thoughts into an elementary piece of code. Both knew that their accomplishments were modest. "[M]uch further experimental and theoretical work remains to be done," acknowledged Shannon.
Unlike most of the early applications of the computer, the general public could understand chess. Many individuals were familiar with the game and knew that it was difficult. They were more impressed with a machine that played chess than with one that solved partial differential equations.
Shortly after the publication of Shannon's paper, Computer Research Corporation, a small California computer manufacturer, learned that if chess could command the public attention it could also overpower any other discussion of the computer. One of the company's executives, Richard Sprague, grabbed onto the idea of using a chess-playing machine in an attempt to explain the computer to a bored reporter. The "power of the stored program idea was so great," he claimed, "that eventually computers of this type would be able to play a decent game of chess."
Sprague's dramatic claim quickly generated a series of newspaper stories. Much to his dismay, the stories were less interested in the capability of his machine than they were in a possible chess tournament among the new computers. The calls for a tournament embarrassed Sprague, as his firm had no plans to write a chess program for the machine.
As the stories began to draw the attention of the New York media, Sprague had to disentangle himself without admitting that he had exaggerated his company's technical ability. "To avoid the admission that the machine could never play chess," he recalled, "we stated that the computer was scheduled for far more important defense work at Cambridge Air Force Research Laboratories. Thereby we graciously ducked the matter."
Chess remained a central focus of computer science until 1996, when Deep Blue became the first computer to beat a grand master, Gary Kasparov, in tournament play. During the intervening years, the development of chess programs created new forms of searching, new ways of organizing memory, and new hardware architectures. However, even with all these accomplishments, this research had little influence on the modern video game. Video game technology came from researchers who were attempting to simulate the social world with a computer.
Unlike computer chess, computer simulation has never captured the imagination of either the public or computer scientists. It "was not then and is not now part of the mainstream of either computer theory or applications," lamented Julian Reitman, a pioneer in the field. Computer simulation developed as an engineering tool, providing a means of understanding complex systems.
The airline industry was one of the early users of computer simulation. In 1955, the airlines formed only a small part of the transportation infrastructure, but they were growing rapidly. American Airlines, the largest US-based carrier at that time, had only 85 planes but it had a grand vision. "Any employee who can't see the day when we will have 1,000 planes," argued the company's president, "had better look for a job somewhere else."
The engineers at American Airlines wanted to use simulation to understand a key part of the organization: the reservation system. This large, complicated system involved reservation agents, clerks, phone lines, airports, travel agents, and paper records. At the heart of the system was the Reservisor, an electromechanical device that counted the number of tickets sold for each flight. The engineers had tried to analyze the system but had concluded that standard mathematical methods did not accurately capture the flow of information from person to person to machine.
The first simulation of this reservation system was neither glamorous nor an event that might draw public attention. "For an entire week," recalled one young engineer, "the table of random numbers was manually consulted, transactions were generated, and a printing calculator was used to perform the mathematical operations." The work was hard because it was repetitious and demanded concentration on details.
The engineers tried to translate their work into a computer program for the IBM 650 but were thwarted in their efforts. It was "beyond the state of the art in the late 1950s," the engineers ruefully acknowledged (J. Reitman, "How the Hardware and Software World of 1967 Conspired (Interacted?) to Produce the First in the Series of Winter Simulation Conferences," Proc. 1992 Winter Simulation Conf.).
Step by step, engineers began adopting the tools of computer science for such simulations. In an early successful example in 1957, traffic engineers at the University of Michigan wrote a computer program to simulate the movement of automobiles at an intersection connecting four streets."[E]very quarter of a second of problem time, the contents of the computer were scanned," reported the research team. To analyze the simulation, they developed a primitive form of computer graphics. On an oscilloscope, they drew points "corresponding to those of cars within the computer model." They then photographed the oscilloscope with a movie camera. "The resulting film provided an animated cartoon-like version of the flow of traffic within the computer." (H. Goode and W. True, "Simulation and Display of Four Inter-Related Vehicular Traffic Intersections," Proc. ACM, 1959, pp. 65-1 to 65-2.)
Like the American Airlines engineers, the Michigan researchers never completed their project. "Funds were not forthcoming," they claimed. Yet, they established the pattern that other researchers in simulation would follow. They worked unobtrusively. They concentrated on analyzing their simulation. They used the newest contributions of computer science. Computer simulations were some of the early applications for innovations such as object-oriented programming, interactive graphics, and parallel and distributed processing.
Computer simulation technology caught public attention only when it became the technology of computer games. This transition occurred in the 1970s, when the cost of computation fell. When computers were expensive, they had to be used for important tasks like fundamental engineering, basic science, or corporate operations. As they became commonplace, they could be used as a source of entertainment.
The early games of this period directly simulated physical reality. A well-circulated Basic program of the 1970s simulated a landing on the moon. The classic game of Pong simulated the physical game of table tennis. Simple though they were, these games could captivate an individual for hours. Many an enthusiast spent long evenings hunched over a teletype terminal while playing with the lunar lander program, rejoicing with each success and cursing with each failure.
The games of the early 1970s were often ill-conceived and badly written, but by the middle of the decade, the best of these programs were borrowing the technology of computer simulation. As more programmers became skilled at writing games with simulation tools, they began to realize that they could explore fantasy worlds.
"The computer is a playing field on which one may play out any game one can imagine," wrote one commentator of the time. "One may create worlds in which there is not gravity" or situations "in which time dances forward and backward," or societies that "obey alternative laws of economics." "In short," he argued, "one can single-handedly write and produce plays in a theatre that admits to no limitations."(J. Weisenbaum, Computer Power and Human Reason, Freeman, 1976, p. 113.)
The computer game industry began in 1977, and it continued to grow as if it faced few, if any, limitations. As it has grown, it has used simulation technology to create a host of artificial and imaginary worlds. It has given us artificial air flight, spelunking, warfare, city planning, and environments that have little parallel in the natural world.
We have now reached a point where many, if not most, people get their first exposure to computers through video games. Interactive computer games are a major sector within both the high technology and entertainment industries. They have become the "killer app for multimedia" according to Nolan Bushnell, a founder of Atari, the game manufacturer.
I was unable to identify the artificial world that captured my neighbors' attention. From their body movements, I suspect that the two young men were playing something that simulates physical activity, such as a computerized football game or virtual motocross.
No matter what game these young men were playing, they were using the common technologies of computer simulation. Within the program, there are event-driven actions, pseudorandom number generators, and various forms of objects. By itself, the technology would not occupy them for a minute. However, when that technology was used in a competitive, interactive story, it commanded their attention indefinitely. Doing so can cause players to forgo sleep, curse at inanimate objects, and leave their windows open so that all of their neighbors can observe how a simple computer program can capture the imagination.
David Alan Grier is the editor in chief, IEEE Annals of the History of Computing, and the author of When Computers Were Human (Princeton University Press, 2005). Grier is an associate professor in the Center for International Science and Technology Policy at the George Washington University. Contact him at firstname.lastname@example.org.