The Birth of the Microchip
The electronics revolution was built on developments in three distinct fields: human-computer interface, which connected people to their machines; networking, which connected the machines together; and, most importantly, microchips – the machines themselves. The U.S. government, and more specifically, the Defense Department’s Advanced Research Projects Agency (DARPA), unquestionably played the decisive role in the development of the first two fields. The history of the microchip, however, which powers everything from computers to cell phones to CD players, contained many contributions from both government and industry, here and abroad. In fact, the microchip arguably began in the laboratory of the newly-formed Texas Instruments with the creation of the first integrated circuit.
The concept of the integrated circuit had been around for several years when Jack Kilby was hired by Texas Instruments. The idea was to pack multiple transistors, the devices that regulate current in a circuit, onto a single, tiny chip, which would nearly eliminate the distance between the circuits and therefore dramatically increase the speed with which electronic instructions flow through a computer. Many engineers and scientists were working on the problem, but because Kilby began his new job over the summer, many of his colleagues were away on vacation. So, Kilby worked alone through the summer, and when his colleagues returned, he presented them with the first working prototype. That chip, which contained tens of transistors, began the race to pack more transistors on to even smaller chips – a race that will continue well into the future.
At the time, however, relatively few organizations had the resources to compete, which slowed the pace of progress. University scientists and their students, in particular, had no way of testing their designs or conforming to the different standards of each manufacturer. Just as with the development of the Internet, these proprietary standards forced engineers to duplicate each other’s work rather than profit from it. That began to change when Caltech’s Carver Mead laid down a standard set of design rules, creating a systematic science of chip design and ensuring that new ideas could be easily implemented. Disseminating that standard fell to the government.
In the late 1970’s, DARPA created a program called MOSIS (Metal Oxide Semiconductor Implementation System) that would allow individual researchers and students to test new chip ideas based on Mead’s design rules. Over the next decade, MOSIS brought thousands of new designs to life, which undoubtedly contributed to the rapid increase in computing power and the U.S. domination of the industry. The success of MOSIS highlights one role of the federal government in technology development – ensuring that as many groups as possible have the tools to advance the field. The government’s subsequent contributions to the development of the microchip illustrate another; its ability to turn a competitive marketplace into a collective enterprise.
While U.S. chipmakers produced incremental improvements in size and speed, the Japanese government spent hundreds of millions of dollars on entirely new chip designs, so that by the mid-1980’s, Japanese companies captured the majority of the market and the U.S. industry seemed to be on its way out. The U.S. government responded with three initiatives. The first two, the Very High Speed Integrated Circuit (VHSIC) program and Very Large Scale Integration (VLSI), were funded by the Department of Defense and DARPA respectively, and helped U.S. companies catch up to Japanese technology. The third, Sematech, represented one of the most successful government-industry collaborations. With joint funding from ARPA and many leading chipmakers, Sematech advanced U.S. chipmakers past Japanese competition and saved a $150 billion dollar industry. Both the private sector and the government contributed to the development of the microchip, but the continued existence of a U.S. industry owes a great debt to the federal government.
Malone, Michael S. The Microprocessor: A Biography. Santa Clara, CA: Telos, 1995.
Rowland, Alex and Shiman, Philip. Strategic Computing: DARPA and the Quest for Machine Intelligence. Cambridge, MA: MIT Press, 2002.