Did the government really invent the Internet?
Earlier this month, President Obama argued that wealthy business people owe some of their success to the government's investment in education and basic infrastructure. He cited roads, bridges, and schools. Then he singled out the most clear-cut example of how government investment can spark huge business opportunities: the Internet.
"The Internet didn't get invented on its own," Obama said. "Government research created the Internet so that all the companies could make money off the Internet."
Until recently this wouldn't have been a controversial statement. Everyone in the tech world knows that the Internet got its start in the 1960s, when a team of computing pioneers at the Pentagon's Advanced Research Projects Agency designed and deployed ARPANET, the first computer network that used "packet switching" -- a communications system that splits up data and sends it across multiple paths toward its destination, which is the basic design of today's Internet. According to most accounts, researchers working on ARPANET created many of the Internet's defining features, including TCP/IP, the protocol on which today's network operates. In the 1980s, they strung together various government and university networks together using TCP/IP -- thus creating a single worldwide network, the Internet.
Suddenly, though, the government's role in the Internet's creation is being cast into doubt. "It's an urban legend that the government launched the Internet," Gordon Crovitz, the former publisher of The Wall Street Journal, argued in a widely linked Journal op-ed. Instead, Crovitz believes that "full credit" for the Internet's creation ought to go to Xerox, whose Silicon Valley research facility, Xerox PARC, created the Ethernet networking standard as well as the first graphical computer (famously the inspiration for Apple's Mac). According to Crovitz, not only did the government not create the Internet, it slowed its arrival -- that researchers were hassled by "bureaucrats" who stymied the network's success.
"It's important to understand the history of the Internet because it's too often wrongly cited to justify big government," Crovitz says. I'll give him one thing: It is important to understand the history of the Internet. Too bad he doesn't seem interested in doing so.
Crovitz's entire yarn is almost hysterically false. He gets basic history wrong, he gets the Internet's defining technologies wrong, and, most importantly, he misses the important interplay between public and private funds that has been necessary for all great modern technological advances.
If you spend time looking at the history of the Internet, you'll find the government there at every step. Researchers working directly for the government and at university labs funded by the government were some of the first people on the planet to think up a worldwide network, and, at the beginning, they were the only people working to build such an outlandish thing. That's not true just of the Internet. Pop open your smartphone and you'll find government research at the heart of just about every component, from the batteries to the GPS chip to the microprocessor to the multitouch interface.
This doesn't mean that the government deserves all credit for creating your phone. But it does mean that President Obama was right -- in tech, no one does anything on his own. Useful products are usually the result of years of research by smart people at various institutions: government labs, university labs, and corporate R&D campuses. The history of the Internet, like much of everything else that makes our world so magical, proves that in the tech industry, it takes a village.
If you want to find out who built the Internet and why, there are a few main sources you should consult. If you've got time, read "Where Wizards Stay Up Late," Katie Hafner and Matthew Lyon's definitive history of the founding of the worldwide network. If you don't have much time, look at "A Brief History of the Internet," written by many of the scientists who worked on the system in its early days. The many Wikipedia articles on the history of the Internet are also quite helpful. All these sources put the lie to Crovitz's ridiculously partisan theory that Xerox, and not the government, created the Internet.
Some of Crovitz's errors seem to stem from technological ignorance; in arguing that Xerox's graphical machines were in some way responsible for the design of the Internet, Crovitz seems to conflate the Internet and the World Wide Web. The Web is the system of linked, usually graphical documents you see in a Web browser -- i. e., sites like Slate. The Internet is the network over which the Web and other communications systems -- email, instant messaging, file-sharing -- travel. The Internet predated the Web.
Other times, Crovitz strays into what seems like intentional intellectual dishonesty. He mentions offhandedly that "Vinton Cerf developed the TCP/IP protocol," but he skips over both the gravity of this development and the government's role in it. TCP/IP is the Internet's defining language, the only reason that any two computers, anywhere, can send a message to one another. In this way, TCP/IP is the Internet. What's more, Crovitz neglects to mention that when Cerf created TCP/IP, he did so with Robert Kahn, who was an employee of the Defense Department, and that both of them were working under funding from the government.*
What about Ethernet, Xerox's networking system, which Crovitz credits with being primarily responsible for the Internet? He's right that Ethernet certainly was vital to the widespread adoption of networked computers. But as Ars Technica's Timothy Lee points out, Ethernet connects different computers into a single network and does not connect different networks into a single larger network -- which is what the Internet is, by definition. Calling Ethernet the foundation of the Internet is like calling sidewalks the foundation of the modern transportation network. Sure, sidewalks are important in how you walk around your neighborhood, but they don't play much of a role in how you get from Paris to New York.
Crovitz's contention that the government slowed down the Internet is also totally backward. In fact, if you want to blame any single institution for delaying the Internet, your best bet would be the largest private corporation in the land -- AT&T. In 1960, an engineer named Paul Baran came up with the idea of a packet-switching network. Baran was working for the RAND Corporation, a government-funded think tank, and he'd been looking for ways to create networks that would survive a disaster. Baran saw that the country's most basic communications infrastructure -- especially the telephone network maintained by AT&T -- had several central points of failure. If you took out these central machines, the entire network would fail. His insight was to create a decentralized network, one in which every point was connected to every other point in multiple ways -- your message from New York to San Francisco would get split into packets and might pass through Chicago, New Orleans, Atlanta, Tampa, or St. Louis. If one of those nodes were taken out, most of your message would get through, and the network would still survive.
As recounted in "Where Wizards Stay Up Late," when Baran presented this idea to AT&T, the company's engineers thought he was nuts. They argued that he had no idea how to run a communications system, and they fiercely resisted the idea of creating a packet-switching network. And that's why the task fell to the federal government -- the Defense Department had to create the Internet because private enterprise refused to.
Decades later, it's easy to blame AT&T for being shortsighted. At the time, though, the company's decision seemed perfectly reasonable. Baran was proposing something completely radical -- who in his right mind would route a message from New York to San Francisco through so many different paths? And why make such a huge change when AT&T's old way worked so well for its own aims (that is, building a profitable business)?
In other words, creating something as grand and untested as the Internet was something that a private company simply couldn't do. The project was too big, and the payoff too uncertain. That's true of most technologies in their infancy. The Army created ENIAC, the world's first general-purpose computer -- and only after the military proved the basic idea was sound did IBM jump into the business. Apple began working on a multitouch interface in the 2000s, but that was only after decades of research at other labs, including by many researchers funded by the government. The American military developed and launched the network of satellites that form the Global Positioning System -- and only then could tech companies come along to make spectacular use of that system.
None of this is to deny the importance of private enterprise in tech. Apple, Facebook, Amazon, Google, Microsoft, and every startup in Silicon Valley deserves credit for today's tech marvels. But not a single one of them could have done much of anything without pioneering work by the government. The Internet, the Web, the microprocessor, GPS, batteries, the electric grid -- if you've built a thriving company that depends on any of these things, you didn't get there on your own. Or, as the president once said: You didn't build that.
• Farhad Manjoo is Slate's technology columnist.
- Share Facebook Twitter
Article sent to (required)E-mail
Article sent from (required)E-mail Name
Subject Line (article title)
Message (optional)Success - Article sent Click to close
Interested in reusing this article?
Custom reprints are a powerful and strategic way to share your article with customers, employees and prospects.
The YGS Group provides digital and printed reprint services for Daily Herald. Complete the form to the right and a reprint consultant will contact you to discuss how you can reuse this article.Need more information about reprints? Visit our Reprints Section for more details.
Contact information ( * required )Name * Company Telephone * E-mail *
Article InformationTitle URL
Message (optional)Success - Reprint request sent Click to close