A history lesson on government R&D, Part 1

When President Obama gave his State of the Union address on January 25, 2011, I noticed he was criticized specifically for his so-called “Sputnik moment,” his call for “investment” in certain areas involving science and engineering. Here is some of what he said on that:

Half a century ago, when the Soviets beat us into space with the launch of a satellite called Sputnik, we had no idea how we would beat them to the moon. The science wasn’t even there yet. NASA didn’t exist. But after investing in better research and education, we didn’t just surpass the Soviets; we unleashed a wave of innovation that created new industries and millions of new jobs.

This is our generation’s Sputnik moment. Two years ago, I said that we needed to reach a level of research and development we haven’t seen since the height of the Space Race. And in a few weeks, I will be sending a budget to Congress that helps us meet that goal. We’ll invest in biomedical research, information technology, and especially clean energy technology, an investment that will strengthen our security, protect our planet, and create countless new jobs for our people.

People within the computer science, science, and engineering communities have been pining for a “Sputnik moment” for several years now, hoping that something, anything, will inspire our society to make math, science, engineering, and computer science a higher priority, and promote more funding for those fields in academia. It’s not just academics who are worried about this. The industries that count on having a supply of good scientists, engineers, and mathematicians to draw from in the U.S. are worried about the decline in interest among American students as well.

The hope for this “moment” is misguided, in my opinion, because what they’re really hoping for is something contrived. It also speaks to a lack of faith in America, that we cannot be inspired to pursue these fields without some foreign enemy that we feel we have to compete against. Obama kind of tried to create this “Sputnik moment” in his speech, talking about how the Chinese and South Koreans have better technology than we do. This is true. In the case of South Korea and Singapore, it’s been true for almost ten years. Back in 2002, I was hearing about how in Singapore, people were reserving movie tickets for specific movies at specific theaters, and carrying out cash and credit card transactions at retail outlets, with their cell phones. All they needed to enter on their phone was their PIN. You can watch a Computer Chronicles episode where they talk about this. A couple years later I heard about how South Koreans had higher speed internet service than we did, back when the fastest broadband service to which most American consumers had access was probably 1.5 megabits per second. Koreans were already doing HD video conferencing as a matter of course. It’s come to this country just recently. I’m not convinced, though, that Americans are going to commit themselves to spending years studying technical subjects just to beat the Chinese and Koreans over issues like this.

Sputnik was a cultural realization in 1957 about the implications of the Soviets, our Cold War adversary, getting ahead of the U.S. in space technology, primarily out of an anti-communist sentiment, and concern for national security. The government didn’t have to tell people to be worried about this. In fact, it was the citizenry that was banging on the doors of the government, demanding, “Why didn’t you see this coming,” and, “Why weren’t we first?” The Soviets had developed their own nuclear weapon in 1949, and now they had the ability to launch things into space. People made the connection that “they could drop nuclear bombs on us from space,” a strategic high ground that we were not even close to occupying at the time. Our own test rockets were blowing up on the launch pad on a regular basis. The feeling was the Soviets had beaten us in a game of one-upsmanship, when we least expected it, and there was an alarming sense that we needed to “catch up.”

Edit June/July 2019: To briefly explain what happened, I’m using production clips from a filmmaker named David Hoffman, who has generously released much historical footage for public viewing.

According to what we know from history, the so-called “missile gap” was a political myth. Nevertheless, something good was able to come out of this. Math and science, which our culture had largely ignored and neglected, became a higher priority. We produced more scientists and engineers. Some of them went into university research. Others went into private industry. A couple by-products of this was a heightened interest in getting more women into computer science in the 1970s–an effort that actually succeeded at the time, but only temporarily. It also spurred the explosion in commercial personal computers in the late 1970s, which continued into the 1990s. There was a cultural understanding which said that learning about computers at a technical level, and how to use them, was important for our future.

That culture has changed. We value computer technology, because now it’s everywhere, and it’s hard not to learn about it, but now it’s all about learning the technology’s interface, not how to build it, or manipulate its internals. There is much less interest in what makes our technology possible, and in learning what’s necessary to drive it forward.

The criticism of Obama’s speech where he talks about our “Sputnik moment,” has bothered me somewhat, because the critics missed the history of government research funding. We really shouldn’t dismiss it, and let ourselves be ignorant of it, because I think this is one of the relatively few areas where government has done a good job, even with the occasional misguided notions that have been promoted through it under the label of “science.”

Here is some of what was said against Obama’s speech.

From Investors.com, published by Investor’s Business Daily, “Obama’s Tribute to Big Government”:

Are you impressed with the Internet? With your iPad? With that gadget on your car’s dashboard that gets you back on the Interstate after you get lost in a strange city?

Thank Washington, because according to the president Washington “planted the seeds for the Internet. That’s what helped make possible things like computer chips and GPS. Just think of all the good jobs — from manufacturing to retail — that have come from these breakthroughs.”

If you don’t remember Presidents Carter or Reagan or Clinton bragging about the initiatives of their administrations that would one day bear the fruit of global social networks like Facebook, handheld communication devices like BlackBerrys, and the mobility revolution of Wi-Fi, it’s because they didn’t.

Government didn’t bring us any of those things; the private sector did. Does anyone really believe the federal government — which can’t find, let alone police, the 13 million-plus illegal aliens within our borders — can find and finance entrepreneurs like Bill Gates and Steve Jobs while they are in the embryonic stages of their careers and make them successes?

The part of Obama’s speech that was quoted makes it sound like he was making a bit of a non-sequitor. Here is the full quote:

Our free enterprise system is what drives innovation. But because it’s not always profitable for companies to invest in basic research, throughout our history, our government has provided cutting-edge scientists and inventors with the support that they need. That’s what planted the seeds for the Internet. That’s what helped make possible things like computer chips and GPS. Just think of all the good jobs — from manufacturing to retail — that have come from these breakthroughs.

On these points, Obama was right, but it takes an understanding of the history of this technology to get that. Defense spending was an important part of establishing the internet (through ARPA). It should be noted that the internet I refer to here is not the web, but rather the basic hardware and software infrastructure that the web uses in order to work. NASA provided R&D funding for the development of computer chips around the time they were invented, and without defense spending, there would not be a GPS satellite system, which is essential for GPS technology to work here on the ground. It should be noted that the technology for GPS was initially used exclusively by the military, the technology for which was kept secret, and I think was only made available for consumer use within the last 10+ years. These have been essential technology platforms for the consumer products and services we use.

Just a side note: The example of the level of illegal immigration exhibiting government incompetence is a bad one. It’s not that our government can’t police and regulate this better. It’s that it doesn’t want to, and there are interests in this country who like it that way. Anyway…

There was also this from Glenn Beck shortly after the speech:

The President also announced he would be “investing” in biomedical research. Can anyone find that one in the Constitution? Oh, and information technology, especially clean energy technology. Alright. Let’s take these apart one at a time, shall we?

Biomedical: Maybe it’s just me as an American citizen, but I don’t want the government now dabbling with creating our drugs. Period.

Information technology? I dunno. I thought we were doing pretty good with Steve Jobs. Bill Gates [is] doing good, you know. Newcomers [are] coming into that field all the time. I think we’re good! Have you seen the iPhone…next to the Post Office? I’m going to go with Apple on that one.

After seeing these comebacks from conservatives, I shook my head a bit in disgust. To me, these people clearly didn’t understand what they were talking about with regard to how the products we see today came to be. Well, I hope to remedy that a bit in this series.

Obama wasn’t talking about starting a government-owned biomedical or IT company, or the government becoming a venture capitalist that would fund startups (though, as you’ll see later in this series, the government did that in a round-about way during the “middle stage” of the internet’s development), but rather that the government would provide funding for basic scientific research of the same type that developed the technology platforms which Facebook, Twitter, GPS devices, smartphones, and PCs were able to build upon, or were able to use. It’s not a stretch at all to say that most of these platforms would not exist today, or would’ve come later, had government-financed research not delved into creating them.

I will use several sources for this series. The primary one is “The Dream Machine,” by M. Mitchell Waldrop. He tells the story of where the basis–the ideas and infrastructure–for a lot of the technology we use today came from. Other sources I will use are Wikipedia, historical videos I’ve found on the internet, and a documentary called “The Machine That Changed The World.” Occasionally, I will cite anecdotes I have heard from people who were either involved in this work, or who have been close to the people who were.

The section of “The Dream Machine” that I use follows the career of one man, J.C.R. Licklider, and several other computing luminaries, who made great contributions to the field of computer science, and particularly to the goal of creating interactive, networked computing, the kind of computing we do every day when we use technology.

What follows is a series of posts on this subject. I’ve broken it up into parts, because it’s extensive. The goal of what I’m writing here is not to be an authoritative source, but to provide a primer which you, the reader, can use to further your own research of this topic, if you feel so inclined. I will not cover the full history of computer technology (though there is still a lot here). What I will cover is the history of government-funded research into certain computer technologies, and what flowed from that research. I’ve made a point to try as much as I can to include the contributions of private companies that were created from government projects, or which worked with government agencies, or were beneficiaries of ideas generated by government-sponsored research, and which were critical in bringing us the technology we use today.

Getting computing off the ground

The first technological breakthrough described in the “The Dream Machine” was ENIAC (the Electronic Numerical Integrator And Computer), the first general-purpose computer. There were other computers invented by other creators before this, but they were designed for a specific purpose. It was either impossible, or very difficult, to program them for other purposes. This computer could be programmed to use a variety of calculation methods.

J. Presper Eckert

ENIAC was designed by J. Presper Eckert, and John Mauchly (pronounced “mock-lee”). It was a digital computer, and its electronics were made up of vacuum tubes, also known as “valves.” It was a government project, financed by the U.S. Army, at the Moore School of Engineering at the University of Pennsylvania, during WW II. It was completed in 1946, and took up a whole room. It was also the first high profile project where women were important to the operation of a computer. They were called “Rosies,” and they were the programmers of ENIAC. This was no mean feat for anyone to do, since no programming language existed. It was programmed via. panels of “switches” (though the controls were actually dials), and the only code the programmers had to read were wiring diagrams. Before this, the “Rosies” had worked as human computers, figuring firing tables for artillery shells for the war effort by hand, with the help of mechanical calculators. Human computation was a problem, though. Errors were inevitable, and they might go undetected. It took a long time for people to compute tables. The sense of the time was they needed the firing tables yesterday. They couldn’t produce them fast enough. So building an electronic computer to do this work was seen as essential. As was the case in all sorts of male-dominated fields of the day, since most of the young men in the country were off fighting the war, women were brought in to do both kinds of work (computation and programming), and they showed themselves to be up to the task, though their contribution went unrecognized by society for many years.

John Mauchly

There’s a new documentary out on DVD now called “Top Secret Rosies: The Female Computers of World War II” (h/t to Mark Guzdial) for anyone who’s interested in learning more about this. There’s also a segment on this history in the documentary, “The Machine That Changed The World,” in its first episode, called “Great Brains.”

The following video of ENIAC, and the people who operated it, is National Archives footage digitized to the web. Note: There is no sound in this video. Eckert and Mauchly appear in the video at 4:55.

The blinking light display on the ENIAC was set up solely for the computer’s public unveiling. The display is a panel of lights covered with ping pong balls, with numbers painted on them. This panel was not used for normal operation. Nevertheless, it established the precedent of computers having panels of blinking lights.

There were two limitations with ENIAC. One was, while it could be programmed, it could not store its program. Programming was a matter of “wiring” the program into the machine, which was accomplished by adjusting the aforementioned panels of dials. Second, it was not interactive. The computer loaded data into itself, processed it, and produced printed tables of calculations. That was it. This would more or less become the template for most of the computers that would be produced and used for the next 25 years.

Taking what they had learned from building ENIAC, Eckert and Mauchly formed the world’s first computer company, the Electronic Control Company, later called the Eckert-Mauchly Computer Corporation, in 1946. They developed the Univac (the Universal Automatic Computer). The following promotional film for Univac was produced after the Eckert and Mauchly company was bought by Remington-Rand in 1950.

Univac was made famous when it was featured on a CBS News broadcast, giving the first statistical computer prediction for an election, in the 1952 presidential race. The prediction was based on a sample of actual vote tallies on election night, and was very close to the actual result, though CBS did not air the prediction, because it differed significantly from polls taken before the election. The Univac’s prediction was correct, though: Eisenhower won.

IBM decided to get into the computer business after hearing requests for computers from the U.S. military during the Korean War. There was resistance within IBM to getting into this new market. Nevertheless, it produced its first computer model, the 701, in 1953. Its design was based on Dr. John von Neumann’s paper, “Preliminary Discussion of the Logical Design of an Electronic Computing Instrument,” which documented his work in creating a computer that could store its program, at the Institute for Advanced Studies, at Princeton.

IBM came to dominate the field with its superior marketing, and backward compatibility with the punch card system of its older mechanical tabulation machines.

The first glimmers of interactive computing

What I covered above is one path that computers took, which came to dominate computer use for a long time. Another path that was very significant, though was not enjoyed by the public at the time, was interactive computing. For many years, interactive computing would only be known in the military. Eventually it would come to be known by the whole world.

Jay Forrester, from Wikipedia

Going back in time to before ENIAC was created, the Navy wanted to build a flight simulator for pilot training. They funded a project at MIT called Whirlwind, starting in 1944. It was designed by Jay Forrester. The system he ultimately built was the first real-time digital computer. Like ENIAC, it used vacuum tubes. What was different was it gave constant feedback to an operator (the pilot trainee) as it received input (movements of a joystick by the trainee), and it gave the operator something to look at, a simple graphical display, generated using an oscilloscope.

Whirlwind was never really “finished,” though it was put into military service in 1951. The scientists and engineers who worked on it were constantly tinkering with it, and changing its purpose. Under “Project Claude,” Whirlwind was tasked with displaying the positions of a real drone and a real aircraft flying in the sky, taking radar data as input in real time, allowing the operator to use the joystick that was part of the computer’s setup to remotely control the drone. The goal was to see if the person sitting at the computer could use the equipment to pilot the drone to intercept the aircraft. In tests it worked. This is not unlike the way military drones are piloted today.

A government-deputized committee that was formed shortly after the Soviets tested their first nuclear bomb in 1949, headed by MIT physicist George Valley, saw Whirlwind in action in 1950. It was a decisive proof-of-concept for them about what computers were capable of, and how they could be used for air defense. The Valley committee was tasked with assessing our existing defenses in the face of a Soviet nuclear threat, and making recommendations about any changes that were needed. They found that our defenses were woefully inadequate for defending against a bomber raid, the only major threat that was imagined at the time.

The committee’s report was delivered to the Pentagon in 1950. It recommended that the government install more radar stations to fill in the gaps in the radar net, and that an automated radar tracking and response system–a computer system–was needed. Seeing Whirlwind in action gave them confidence that this was possible. The U.S. government was very nervous about the situation with the Soviets. The thinking was the cold war could turn into a hot war quickly, and so they accepted the Valley committee’s recommendations without question, even though this was a radical, new idea at the time, only supported by experimental technology.

A man by the name of J.C.R. Licklider comes into the story here. He was a psychologist, with a background in mathematics and physics, who was very interested in computers. In addition to being a scientist, he was also a mathematician. He is a major figure in the history that follows.

Licklider came to MIT in 1950 to continue his work in psychology. While there, he developed custom-built analog computers to simulate isolated neurological activity in the human brain, as part of his research. He also contributed to the SAGE (Semi-Automatic Ground Environment) project in 1951, working on human factors research for the displays that operators would be working with. He helped determine when and how much information should be displayed to the operators, and what the intensity of the displays should be, so that the operators would be comfortable, and would not be overwhelmed with information, so they could make clear decisions.

SAGE was, in essence, the system that was recommended in the Valley Committee report. It was designed to be a real-time system, like Whirlwind, which would monitor our defensive airspace for Soviet bombers, and our own air force flights. It was developed at MIT, with several corporations contributing to its construction and planning, including IBM, System Development Corporation (which was spun off from RAND Corp. They managed all of the programmers for the SAGE project), Burroughs, and Western Electric.

The programming task for SAGE was very complex. It was estimated they needed 2,000 programmers. There weren’t enough in the whole country to work on it. So the SAGE project undertook an education program to recruit and train new programmers. All comers were invited, men and women, from all walks of life. It was not easy to tell who the best programmers would be. Scientists, engineers, mathematicians, the typical candidates one would think were ideal, were not sure bets. Music teachers tended to be the best candidates. It was found that women were better than men at paying attention to minute details while simultaneously not losing track of the big picture. One of the best programming groups in the project was 80% female.

The Navy lost interest in Whirlwind in 1954, and stopped work on it. The Air Force began construction of SAGE installations that same year. The SAGE system was brought online in 1958. Each installation was huge, due to the fact that, like ENIAC, its electronics were made up of large arrays of vacuum tubes. It took up a few floors of a building at each installation. Each one had two computers, so that one could always be taken down for maintenance if need be. It was networked so that information gathered about detected objects in the sky could be relayed to other stations. The first phone modems were created during this project, for this information transfer.

SAGE was fully built by 1963. The thing was, it was obsolete by this point. ICBMs (Inter-Continental Ballistic Missiles) had been developed, and the nuclear threat situation had become more a matter of nuclear deterrence than strategic defense. SAGE could not deal with ICBMs at all. Still, Whirlwind and SAGE would have a huge impact on the future of computing in the minds of the scientists and engineers who worked on them. SAGE was the first big proof-of-concept that interactive, networked computing was a viable concept. Still, it would take a lot of convincing to get more players in the computer industry to believe in it. Computers were still big and expensive, and computer time was expensive and precious. The idea of interactive computing was seen as pie in the sky by many–a waste of time, and SAGE was the exception that proved the rule that computers were only for data processing, not for interaction.

What came of it

One of MIT’s working groups on the SAGE project was spun off as the MITRE Corporation in 1958. Its purpose was to integrate new weapons systems into SAGE. In addition, in the coming years MITRE would develop the National Airspace System for air traffic control, which is still in use today, and AWACS (the Airborne Warning and Control System) (sources: Wikipedia, The MITRE Digest).

The SAGE project enhanced IBM’s ability to deliver computer systems that could handle big, real-time data processing projects, because their staff had learned how to do it through the experience of working on this project, and by bringing experienced people in from places like MIT. IBM developed SABRE (Semi-Automated Business-Related Environment) for American Airlines. It was a simplified commercial version of SAGE, but was designed for making airline reservations. The first experimental SABRE system went online in 1960. It was the largest real-time commercial data processing network in the world at the time. It was ultimately linked by phone lines to 1,200 teletype terminals across the country.

American made SABRE available to independent travel agents in 1976, and in the year 2000, spun it off as SABRE Holdings, which exists today. It’s divided into four business units: Travelocity (the e-commerce site for people to make their own travel reservations), Sabre Travel Network (a global distribution system providing travel information to agencies, corporations, and travelers), Sabre Airline Solutions (providing airline reservations systems and financial management), and Sabre Hospitality Solutions (providing technology solutions to hotels).

The SAGE system continued to be used by the military until 1984. Fortunately it was never tested in combat.

J. Presper Eckert, a co-inventor of the ENIAC, became an executive with Remington-Rand when the Eckert-Mauchly Computer Corporation was purchased by Rand in 1950. He stayed on through the company’s many transitions. Remington-Rand was purchased by Sperry in 1955, and went by the name of Sperry Rand, and then Sperry, until the company merged with Burroughs to form Unisys in 1986. It is still in existence today. Eckert retired from Unisys in 1989. He died on June 3, 1995.

John Mauchly, co-inventor of the ENIAC, became a founding member, and president, of the Association for Computing Machinery (ACM), and helped found the Society of Industrial and Applied Mathematics (SIAM). He stayed on with Remington-Rand for 10 years, after their company was bought by Rand. He left the company in 1959 to form the consulting firm Mauchly Associates. He later formed another consulting firm called Dynatrend in 1967. He died on January 8, 1980. (sources: Wikipedia, the Association for Computing Machinery, Ohio History Central)

Jay Forrester, creator of Whirlwind, moved to the Sloan School of Management at MIT, and left the development of digital computing for good, in 1956. He created a new field of research called “system dynamics,” which looks at the interactions of objects in dynamic systems, with an emphasis on simulation of those systems. He developed ideas which have led to modern notions of supply chain management. (Update 11/21/2016: He died on November 16, 2016. Source: New York Times)

In Part 2, I describe the efforts that began in the 1960s to bring interactive computing to the masses.

—Mark Miller, https://tekkie.wordpress.com