Download PDF The Bridge: 50th Anniversary Issue January 7, 2021 Volume 50 Issue S This special issue celebrates the 50th year of publication of the NAE’s flagship quarterly with 50 essays looking forward to the next 50 years of innovation in engineering. How will engineering contribute in areas as diverse as space travel, fashion, lasers, solar energy, peace, vaccine development, and equity? The diverse authors and topics give readers much to think about! Inventing the Future Monday, February 22, 2021 Author: Robert W. Lucky There is a quotation about predicting the future attributed to Alan Kay, a pioneer in computer science. “The best way to predict the future,” he said, “is to invent it.” Of course, we engineers do often invent the future, but sometimes not the futures that we had intended. The spirit of Kay’s quote is that we plan the future that then comes about because of our inventions. We do that all the time for next year’s products. But what about the longer term? Can we successfully plan and invent a future 20 years out, or is this an unreasonable goal? There are many examples of success and failure at planning the future. President Kennedy said that we would go to the moon. We did. The Bell System said that we would have Picturephones. We didn’t. In 2000 the National Academy of Engineering published a list of the 20 greatest engineering achievements of the 20th century (www.-greatachievements.org), and in 2008 it identified a prospective list of 14 Grand Challenges for Engineering for the new century (www.engineeringchallenges.org). The first list included such achievements as electrification, the automobile, the airplane, and the internet. The second list, of aspirations, included affordable solar power, power from fusion, and secure cyberspace. All of the achievements in the first list were evolutionary developments over decades, and their social impacts were realized over even longer periods. With the possible exception of laser technologies, none of them was based on a singular invention, but rather on an incremental path of development and acceptance while the world changed in unpredictable ways over the duration of that path. Setting a goal for the future is almost a prediction in itself. I remember with embarrassment my participation in a televised discussion about the future in about 1980. The other participants painted bleak pictures of the future in their respective domains. When it was my turn, as the “technologist,” I said that technology would improve the quality of life in the future—and that “we would have big TV sets.” The others looked at me with derision, which was well deserved. It was a silly and useless prediction, even though it was probably the only prediction on that show that came true. But now I wonder: What should I have said, knowing what has happened since? Perhaps instead of saying we would have big TV sets, I might have said that we would have small telephones. That would have turned out to be a much more significant prediction, and it could have been predicted in 1980. Moore’s law of exponential progress in microelectronics technology had been known since 1965 and the year that it would enable a pocket-sized phone could have been predicted. But here is the rub: in 1980 we didn’t know that we needed small telephones! Much of the future is like this. At that time the internet was in its infancy; its future emergence—at the time completely unforeseen—greatly enhanced the utility of a small telephone. Planning the future may involve a long-term goal, but it is important to have a sustainable pathway with waypoints for possible adaptation and redirection. Sustainability is not just a technical issue but also that of support through funding, usually either government financing for a social or military purpose or industry support based on market mechanisms. However, I am wary of plans or procurement contracts with long-term goals that limit freedom for change as technology and conditions inevitably change. I remember two research projects from the 1970s and ’80s that illustrate some of these issues. One was the development of fiber optic network technology, the other involved open-ended research on neural networks. The fiber optics project continues to this day, with incremental progress that has been incorporated in commercial networks at many points along the way. Relevant inventions have occurred with regularity, including improved fibers and lasers and the Erbium-doped fiber amplifier. In contrast, the neural network project was a case of a technology looking for an applicable problem, and it was very difficult to maintain support at that time. Decades later, there were breakthroughs in mathematical algorithms that, combined with a growing recognition of the importance of large datasets, led to neural networks becoming a centerpiece of the surge in machine learning. For neural networks, their time had come. This may be true of many technologies; there is a time of ripeness, and a time when they are fallow, awaiting an unpredictable breakthrough. The neural network example also illustrates the fundamental change in engineering that has evolved over recent decades. Back in the late portion of the last century, I remember a kind of motto recited by researchers working in the materials research division. “Everything must be made of something,” they said. It’s still true, of course, but not in the way they meant it then. Today, like the neural network, much progress is virtual, inspired by mathematical analysis, implemented as algorithms in software, and even designed by computers themselves. It’s as if things are now actually made of “nothing.” It is, however, an enormously powerful and flexible, though difficult and expensive, nothingness. Although none of the Grand Challenges has yet been attained, the list has aged well and still represents important social aspirations. If 2020 was to be a waypoint for these projects, engineers could assess the progress of each and reassign aims for the next waypoint. For example, fusion power may require breakthroughs, while affordable solar power might arrive from steady engineering improvement of known technologies. The secure cyberspace might never happen, but could be a work in progress as long as there is an internet. So, can engineers predict the future by inventing it? In the short term, yes. In the long term, probably not, but we can nevertheless imagine and aspire to what we might accomplish that would be of benefit to all of humanity and plan accordingly. About the Author:Bob Lucky (NAE) is the retired corporate vice president for research at Telcordia Technologies.