The vision that led to the Internet has existed for a long time. In 1908 Nikola Tesla envisioned a device no larger than a watch that would enable a person to communicate with anyone anywhere in the world. In the 1930s H.G. Wells wrote about a “world brain” that would serve as a depot for human knowledge. In the 1940s Vannevar Bush wrote about the Memex, a machine that would store the collective memories and creations of humanity.
In the 1960s, Leonard Kleinrock, Distinguished Professor of Computer Science at the University of California, Los Angeles, built with his colleagues a computer network that he saw as the forerunner of a system that would be always on, always available, and always accessible to anyone with any device from any location. He and the other developers of the Advanced Research Projects Agency Network, or ARPANET, had a model that they had analyzed, so they knew how it should behave. But “as engineers, we also knew what we didn’t know,” he added. “We had to experiment with this thing.”
They built tools into their model to generate traffic, to test the model, to try to break it, and to learn from the results. “The wonders of engineering come when you try something and you discover things that you didn’t anticipate. Then you find out why they behave in that way and how to fix it. The engineering approach was critical. It is not as if we designed something, built it, and let it go, because it would not have worked. We needed to correct it along the way.”
The culture of the time was one of openness, he recalled. “[The model] was shared. We trusted everybody. We knew everybody that was coming into the network. We didn’t put in any impediment—call that ‘security measures’—to protect against bad behavior. In fact, the
gratification that we received as engineers was not to build a system to make money, but simply to solve the intriguing network engineering problem… . The gratification we got was when other people used the things that we designed and made available.”
Once two nodes of the network were built in 1969, at UCLA and the Stanford Research Institute (SRI), Kleinrock and his colleagues decided to test it. Famous first messages, Kleinrock observed, have included “What hath God wrought,” sent by Samuel Morse over the first long-distance telegraph system; “Mr. Watson, come here, I want to see you,” sent by Alexander Graham Bell to his assistant over the first telephone; and “One giant leap for mankind,” uttered by Neil Armstrong as he stepped onto the moon. “Those guys were smart. They understood public relations,” said Kleinrock. “With ARPANET, all we wanted to do was to log in from the UCLA computer to the SRI computer.” A programmer at UCLA typed the letter l. Someone on the telephone at UCLA asked whether the l had arrived. “I got the l,” came the reply. A few seconds later: “Did you get the o?” “I got the o.” Then the system crashed.
“So the first message on the Internet was ‘lo,’ as in ‘lo and behold,’” said Kleinrock. “We could not have asked for a better, more succinct, more prophetic message. But nobody knows it. Hopefully someday it will be as popular as ‘Mr. Watson, come here, I want to see you.’”
THE DARK SIDE OF THE INTERNET
After 25 years of growth, the first massive spam message on the Internet came from two lawyers named Canter and Siegel, who were trying to sell their services to help people get green cards. Their Internet service provider received so many emailed objections that it crashed. “The response to the first spam message was the first denial of service attack—inadvertently,” Kleinrock noted.
The Internet allows anyone to reach millions of people instantly, easily, anonymously, and at no cost in money or effort. “That is a perfect formula for the dark side of the Internet.”
Today, young people cannot conceive of a time when they could not share their photographs, chat with their friends, stream video, or shop online. But the Internet is a disruptive force, Kleinrock observed. As more information is digitized, as devices get smaller, smarter, faster, and cheaper, and as broadband, wireless, and wired continue to explode, disruptive events will continue to occur.
“The Internet itself is still a teenager, and it is acting like a teenager,” Kleinrock said. “It is mischievous. It is erratic. It is unruly and it is disobedient. The hope is that it will mature and grow up into a properly behaved adult. We are not sure. We hope it will.”
THE FUTURE OF THE INTERNET
In the future, content, technology, applications, services, and function will continue to converge in devices that everyone is carrying with them all the time. “The future will be one of extreme mobility, mass personalization, video addiction, location-based services, surprising apps, and, as we know, dramatic society and lifestyle changes.”
A tipping point has recently occurred, Kleinrock observed. Engineers used to develop technology, and applications would follow. Today, applications are racing ahead of the technology, which is trying to catch up to serve the needs of the applications.
Yet the beauty of the Internet’s design is that it does not preclude unanticipated services, applications, or technologies. It has accom-
modated the personal computer, e-mail, mobile apps, the World Wide Web, peer-to-peer sharing, blogs, social networking, and other applications, many of which have come as a complete surprise. “The technology is adaptive enough and general enough to allow these things to blossom.”
Today, the Internet of things is starting to take shape. Already, much of the traffic on the Internet is generated by software agents and embedded devices. Eventually, sensors, actuators, logic, and memories will disappear into the walls, into eyeglasses, into tables, and even into people’s bodies, said Kleinrock. “This room should be alive with technology. It should know I am here. It should know we are all here. If I have a conversation with Corale, it should assist us in that conversation. It should understand what we are saying and understand our background. The idea of making the environment everywhere aware is one of the major benefits we are going to get.”
In the future, the Internet also will become increasingly invisible, so that people using it do not need to think about it. Human-computer interfaces will be far simpler than they are now, so that people will not have to do complicated typing on tiny keyboards. Computers will understand human speech, and apps will be far simpler and easier to use. In time, the Internet will become a “pervasive global nervous system.”
Teaching Students to Overcome Failure
In response to a question about the education of engineers, Kleinrock said that educators need to encourage curiosity, even if it results in failure. “As teachers and faculty, we lie to our students. We present them with 4,000 years of knowledge crystallized and kernelized and lead them to believe that breakthroughs were easily accomplished. ‘Here is this gem. Here is Shakespeare. Here are Maxwell’s equations… .’ The students then think, ‘Well, it is easy. We will get to the frontier and we will be the next whoever.’ But when they approach the frontiers of knowledge, they realize, ‘My god, I can’t move a thing.’”
He also labeled computers “the worst enemy of deep thinking.” Students run a computer model and think they then know how a system behaves. “They have to do more than computation,” he said. “They have to understand their own results and then try to extend them.”
At the same time, students need to learn that they can make a difference, despite the difficulties they will encounter. “They have to be encouraged and applauded for their successes and encouraged with their failures.”
These developments will not all be positive, said Kleinrock. Engineers often have the attitude that they more or less understand the systems they build, whereas natural scientists are forced to deal with a system they did not create. But as engineered systems become bigger and more complex, they are behaving in ways that engineers cannot necessarily understand or anticipate. “In all aspects of engineering, we have to step back and address complexity as an issue.”
The best prediction about the future of the Internet is that it will be unpredictable, Kleinrock concluded. “When [applications] come, they are usually unanticipated, surprising, and delightful, and they take over very quickly.”