In earlier entries, we told you about services that started in the private sector but ended up in the public because governments could do the work cheaper and better, things like public transit, public schools, and fire safety and prevention. Now, let’s turn to something that began as a public innovation but ended up run by private interests.
It’s the internet.
Yes, the most transformative technology of our lifetime—on a scale of innovation, it ranks somewhere between the internal-combustion engine and electricity—began as a federal government project and was financed almost entirely by the government in its first decade. In fact, it wasn’t until the early 1990s, when the internet was more than 20 years old and other ways of connecting personal and business computers had been tried, that its commercial potential was recognized. When that happened, the internet went . . . well, viral.
Before we dive into its history, though, a quick definition. The internet is a globally distributed network of voluntarily connected computers. If you own a personal computer or smartphone and have an internet connection, you have access to portions of these computers. Once there, you can browse their contents freely.
And what access! You can do research online, read newspapers, watch funny videos, play games with people a continent away, drop in on government meetings, shop for all manner of goods and services, send emails to friends and colleagues, trade documents and spreadsheets, do virtual tours of museums and galleries, pay a visit to your childhood neighborhood using Google Street, watch movies, television and concerts online, post a resume, keep up with college and high school friends . . . and that’s just the beginning.
Great fortunes have been made by the internet. We’ve seen the rise of Amazon, Google, Facebook, Netflix, Ebay, Airbnb, Twitter, Uber, Instacart, Yelp and a thousand more whose business models depend almost totally on internet access. We’ve seen perhaps an equal number vanquished: Toys R Us, Blockbuster Video, Borders Books, and Tower Records, among others. More than 20 percent of local newspapers have disappeared in the past decade and a half, as internet rivals took away the advertising that once supported them.
Since the mid-2000s, the internet has introduced things to our lives that would have been impossible without connections at any time to anyone anywhere. The smartphone and the tablet are the most obvious, but there are others: smart speakers (like the Amazon Echo on your kitchen counter), smart thermostats, smart plugs (they turn lights on and off with voice commands), smart televisions that come equipped for streaming. Oh, and don’t forget Zoom and Skype, and the explosion in television streaming services like Disney Plus and Netflix. Tying everything together are the broadband and wireless connections that make instant access possible.
In short, we’ve come to depend on the internet in our everyday lives. So how was it created? What role in its creation did the federal government play? And why did the government lose control of it?
The last question is easiest to answer: because, after a point, the government had no interest in controlling it. A federal agency created the internet in 1969 to serve a single purpose. It was an accident that, as companies were looking for ways of linking computers in the late 1980s and early 1990s, they discovered the best solution was the one the government had made. And was willing to share. For free.
What was that purpose? To allow university researchers working on federal grants to communicate with each other and share computers. The purpose was so limited, it wasn’t even called “the internet” in its early years (it got that name in 1974 as a shorter version of “internetworking”). It was called ARPANET because it was a project of the federal government’s Advanced Research Projects Agency.
ARPA (later called DARPA when the word “Defense” was added to its name) was started in 1958 after the Soviet Union launched the world’s first satellite, Sputnik. Worried that America was falling behind in applying technology to national defense and space exploration, President Dwight Eisenhower created ARPA as a funding agency to identify and manage these breakthroughs.
ARPA’s projects included things like global positioning systems, nuclear test detection, and advanced radar systems, but the actual work came from scientists and researchers at universities, private labs, and the research and development operations of technology companies. The people at these institutions often needed to work together, but there was no practical way of doing so at a distance.
There was another issue. The computers necessary to do computations for these projects were expensive in the 1960s. Since government grants were financing the machines, ARPA wanted a way of sharing computers rather buying new ones every time a university or private lab joined a project.
That’s why ARPA launched a project to facilitate its own research, which it called computer “time-sharing.” From our perspective in the 21st century, this may seem like an easy problem to solve, but in the mid-1960s, there was nothing easy about it. There were, in fact, a series of knotty issues: How could computers at different universities and labs speak to one another using a common computer language and a standard set of procedures? How could we keep these computers secure? Over what infrastructure could the communications flow; would regular telephone lines work or do we need special lines? What is the best way for computer communications to travel—as voices travel over phone lines or in some other way? How could we initiate the communications (we would call this a keyboard today) and how could we see the results (a monitor)?
By 1969 those questions had been answered, and an attempt at computer time-sharing was set up between a team at the University of California Los Angeles and its counterparts at Stanford University, 350 miles away. That October, a researcher at UCLA carefully typed in the first word ever sent in the internet: “LOGIN.”
Except that he only managed to type “LO.” The system crashed.
Designers worked out the bugs. And ARPANET became a reliable and adaptable system, using computer protocols and other innovations that ARPA financed (a big one was called “TCP/IP,” which dictated how computers would connect to one another using devices known as routers). By the mid-1970s other government contractors, including those working on federal energy projects and the space program, joined the now-renamed internet. Gradually, university researchers outside the U.S. also began connecting their computers using the protocols the federal government had pioneered.
In the early 1980s, another technological revolution occurred: the rise of personal computers. Businesses of all sorts suddenly had access to inexpensive desktop computers, as did a growing number of individuals in their homes. These PCs were great at word processing and spreadsheet analysis, and networking had advanced enough that linking them within a business was not difficult. As a result, you could send emails and spreadsheets to colleagues down the hall.
But it was still difficult to tie into distant databases, and sharing documents with anyone not down the hall often involved a three-step process. Step one: You copied the document to a floppy disk. Step two: You mailed the disk to a friend or colleague. Step three: She opened the mail, put the disk in her computer drive, typed A:, hit the return key and . . . voila.
Clearly a better way of connecting this rapidly expanding universe of personal computers was needed. And private companies responded. In the 1980s and early 1990s, three companies offered ways of doing research online, connecting people with similar hobbies and interests, and exchanging emails, first within their networks, then with people outside their systems. Compuserve, AOL, and Prodigy did this using conventional telephone lines and modems.
But this kind of private networking came at a high price in time and money. In the early 1990s, these services typically charged monthly fees plus an online charge of $1 to $6 an hour. And they were notoriously slow. In 1994 the fastest modems downloaded at 28.8Kbps (translation: 28,800 bits per second), which was acceptable if you were looking only at text. But if you wanted to view something with an image, the best way was to start the download . . . and take a coffee break. (By the way, the average speed of a broadband connection today is 64.17Mbps. That’s more than 2,000 times faster than the best 1994 modems.)
By the mid-1990s, many were tired of the limited, expensive confines of AOL and the others, and a new kind of company appeared, internet-only service providers. These were telephone companies, cable companies, and independent providers like MindSpring and Earthlink. They did not have hobby groups or restricted databases. For a flat fee, they got you on the internet; the rest was up to you.
What users found when they got there was a fast-growing collection of private companies, nonprofits, and governments that had learned they could communicate directly with customers and citizens without going through the private services. And this was possible because, in 1986, the federal government had basically thrown open the door, allowing any organization with a router, a computer to share, a website, and a willingness to follow the protocols to join. This caused an ever increasing number to claim their corners of the internet—including, in July 1995, a start-up company calling itself Amazon. And with that, the internet truly took off.
By that point, the federal government’s support of the internet had more or less ended. Today’s internet is governed in a loose way by international nonprofits that decide things like how uniform resource locators are awarded (these are the letters you type in to visit websites like amazon.com or fromthegovt.com) and how the World Wide Web changes over time. But if the government is no longer a guiding force, neither are the private companies that dominated the computer connection business in the early 1990s.
And it’s worth asking why. Why did this government-created “network of networks” succeed and proprietary ones like AOL and Compuserve fail? What does this tell us about public and private control of infrastructure? And had the private networks succeeded, how different would the online world be today?
Let’s start with the last question. And maybe the best way of thinking about it is to consider shopping malls and downtowns.
Shopping malls are privately owned and, until recent years, were quite successful. They are controlled by their owners, of course, which usually makes them more secure than other public spaces and (some would say) a bit boring. Had the private networks succeeded, then, the online world would be like visiting a mall: more secure (fewer business scams and conspiracy sites and less pornography) and more expensive, with users paying to connect and businesses, nonprofits, and governments paying for a presence. And there would have been less to it. Probably no cat videos, Wikipedia, movie databases, or health advice sites. And no fromthegovt.com. In other words, a little boring.
That’s obviously different than today’s internet, which is open, unlimited, diverse, contentious, and more or less free to use. It’s also seamy and lawless in places. Like a big-city downtown.
And, yet, the freest market we’ll see in our lifetimes, the one comprised of internet users and website developers, chose the downtown-style internet over the shopping mall environments of AOL and the others. Why? Well, it was cheaper, so there were fewer barriers to entry. But probably the most important thing was that the internet, designed by the government to be free and accessible to researchers, was also less controlling. To shift analogies, it was more like a public park, less like a country club.
There was one more thing. There were three large private networks, but only ONE internet. And it turned out that ubiquity was important. For companies it meant one website, viewable by all, not three designed for different networks. For users, this meant if you emailed a friend a link to a funny video or news story, she could open it.
This is “one size fits all,” and sometimes that’s exactly what we want: a shared resource, accessible by all, easily understood, and governed by a single set of rules. And no matter how some sneer at “one size fits all,” it’s the way government usually works and, in the right setting, a core strength. One system of streets and roads in a city, not competing systems. One one set of international trade regulations, one Postal Service, one driver’s license bureau with one set of requirements.
Finally, there’s an important lesson here for governments. And that is that it’s important to let others adapt your solution to their needs. The internet became a revolutionary technology not because it allowed researchers to share computers but because private interests used it in ways no one could have imagined in 1969. The federal government’s admirable response was its willingness to go along—even to the point of giving up control of the system it had created.
This development-and-adaption process is critically important in economic development, and more common than we recognize. During the 2020-21 pandemic we saw an example as cities around the country gave restaurants permission to use street parking for outdoor dining. We think of governments as rigid, bureaucratic institutions. But in numerous instances—including the development of the most transformative innovation of our lifetimes—government has been anything but rigid. And for that, we can give thanks.
Postscript: Did Al Gore “invent” the internet? In 1999, then-Vice President Gore used some unfortunate words in an interview that made it sound as if he had created the internet. He hadn’t. But he was the first federal elected official to recognize the internet’s importance and advocate for its development. That began in the late 1970s, when Gore was a congressman, and he kept up his advocacy for the next two decades as a senator and vice president. So, no, Al Gore did not “invent” the internet, no more than George Washington “invented” America. But if Washington is considered the “father of America,” you could reasonably consider Al Gore the political father of the internet.
Give the credit to: federal government
Photo by OiMax licensed under Creative Commons.