 One cannot speculate for very long about life in the 21st century without mentioning the computer. Since the early 1940s, its amazing development has influenced our discoveries, our progress, our knowledge, and most of all the way we think. By looking ahead to the future of computing, we can glimpse our own future. And in thinking about the machines of tomorrow, we must also ask ourselves questions like, where do we want to be? What kind of society do we want? What do we want to do? What changes must we make? Just as sharing a network amplifies the power and intelligence of individual computers, so too dialogue amplifies and extends human vision. With this in mind, we at the Lewis Research Center offer our hopes and our dreams about the future of computing. By sharing just a few of the many diverse perspectives that exist, we hope to begin a provocative chain of thought that begins with you, the viewer of this dialogue, and ends with a magic of tomorrow. I think everything is going to come down to information and communication. Computers today are about the same stage that the automobile was in about 1915 or so, and the networks are going to have to expand in the same way that the highway system has expanded since then. The most important component of the vision of the future is the emerging information superhighway. This information superhighway will operate at 2.4 gigabits per second. That's 2.4, almost two and a half, thousand million bits per second. This highway will have entrance and exit ramps that will reach into every facet of your life. This highway, unlike the concrete highway that carries your car, will be composed of fiber optics. I just kind of picture that most people will just have kind of the equivalent to a hand calculator that will just patch them into the local network, wherever they happen to be, just because of their presence to it. In the future, I think 30 years from now, you'll see a combination of a telephone and a computer. Probably in 20 or 30 years, everyone will have a PC then, like we carry around credit cards now. Maybe they'll even be that small. I remember looking at the cartoon, the Jetsons, and for some reason I have in my vision that eventually that will actually happen to where computers would be everywhere. Something that's part of you will respond to you, will respond to voice commands, will respond to motion, will respond to video. It will be an extension of you as a human being. It'll just be all around us and we won't even think about it anymore. I think the classroom of the future won't be in one place. We've all heard about distance learning. In most aspects, this is still in pilot project form. I imagine with the superhighway, my vision is, it will be part of our way of doing business. As we are able to provide remote communication and remote methods of access and much more user-friendly access, the librarian will no longer be needed to identify what types of information to collect, but more they will be needed to collect information beyond the factual level, towards the knowledge level. At OAI, one of the things that we've started to do is to put in a television classroom at each of our member universities. Distances will disappear. We are linking together students here in the workplace at NASA Lewis Research Center and instructors and students at universities all over the state. If you have team members who are located geographically separated distances and can be connected so that they really feel like they are together, it's got to help. It's just got to add to this ability. I demand you will form dialogue groups well beyond anything the group philosopher Plato ever envisioned as he lectured his student Aristotle. I think you see in the world today that there's more of a common global understanding and maybe a lot of the barriers that we have, international barriers, will begin to melt, at least I hope so. The ability to design a system that's closer to the edge is going to free us up with the computer freeing up the engineer in the sense that his wildest thoughts can be tested out in a sense of reality will free him almost from the rigid constraints of the past. You may have a theory, you may have an equation, a formula that tells you how something is supposed to behave, but most people I know can't visualize formulas very effectively. It would be very nice to be able to have a system for example a truly intelligent computer system that allows you to interact with a computer capabilities that you have both symbolic and numeric and graphic on a high level. It is what you might want to really like to do is go into a rotating engine if you could in a numerical model and perhaps using a glove or your body walking through the engine and positioning through it to see a graphical image of whether it's pressure data, temperature data, accelerometers, velocimeters, displacement transducers whatever phenomenon, physical phenomena you're trying to see to see a graphical image or video image of that data as you would move around through the engine. I envision some kind of holographic type workstation where you'll have an imaging setup where you'll actually see engineering models in a three-dimensional form and that might change a little bit exactly instead of having a monitor you're going to more or less have a stage where a lot of these things will be projected and you might even be able to do tests and run them and see the 3D holographic image move into animation while you're doing some kind of test on it. We're already seeing some changes in that because you can do parametric studies with mathematical models you very seldom do parametric studies in the experiment anymore. You do experiments very carefully with lots of instrumentation and you do a few experiments very very well instead of doing a lots of different experiments where you would just change a parameter slightly and rerun the experiment. You also use mathematical models and computers to predict where the interesting phenomena will occur. I think that the future of computers to predict the performance of liquid chemical rocket engine systems and tire systems from tank to nozzle exit is very promising. And we'll be able to do some amazing technical things at that point. People already are doing some some calculations with the direct numerical simulation of turbulence and transition types of studies for Navier-Stokes equations. You'll be able to do that for chemically reacting flows. You'll be able to do very complex problems. One of the projects that's going on at Lewis right now is this project of trying to simulate an entire jet engine as it's running. We'll have a much better handle I think on methods of failure in materials on things that allow us to design things design objects airplanes spacecraft whatever a lot more cutting edge in the sense that we can make things strong where they need to be strong we don't need to put excess weight where they don't need to be heavy. I do see developing as time goes on and it's starting to happen now is a change in the way we develop algorithms. In particular what I'm talking about first is that if you're going to specify the algorithm explicitly you would be writing a dealing with a program which is writing the program that is programs will be writing programs to the extent that computing can model and help us understand and deal with irrational things fuzzy logic neural network whatever I'm not sure what those will be but I think the challenge for computing is in modeling what goes on in the world computing particularly high performance computing offers us a chance of investigating those concepts. Another thing that it allows us to do which we weren't able to do in the very early days is to work together in teams when we're not living together or at least co-located in teams with high speed networking and with powerful workstations it's possible to carry on a conversation among four or five people right through the computer where ideas can be bounced off each other and tested and with simulation of what's actually going to happen ideas can be tested quickly and that makes it kind of neat it's the ability to do things that not only are going to be cheaper to do now that we couldn't do but in ways we can get information that we couldn't have gotten and so we can make decisions based on more information now with computing than we could have in the past. I have seen computing change from the card program calculator where you had those horrible cards to what we're having now. We put the cards in a tray and it picked up in a truck from the DEB and carried over to the 10x10 where the computer was and then run and then transported back. One of the first big machines we got was the 704 that had 32k words of 36 but words of memory and that now I can get on a pc for about less than $2,000 at that point we paid a good deal of money for that machine and thought we had something really great. We have theories and theories always work far ahead of what we were capable of doing and in less than five years most of the theories were all caught up with in terms of the computer's as of today. I think that we're seeing the seeds of a different type of machine that I think is going to become much more predominant in the future and the seeds I'm referring to is the infant work and neural networks. So what you're looking for is a symbiosis between the number crunching capabilities of current day computers, the qualitative decision making capability of expert system style AI and then the intuition sort of that the neuralness can capture. Then you have something that starts to look like and behave like perhaps like us humans do with one side of our brain being the logical decision maker and the other side bringing the intuition. There's different parts of our brain that specialize in areas. I see a parallel computer doing the same sort of thing. A part of the parallel computer would handle the vision part part would handle the number crunching part it would specialize in different areas. I think we're going to be moving to arena where you won't ask questions relative to the speed of a machine or its memory but how fast can it learn? You could almost send him to training or to school and maybe overnight with computer speeds and they would become a technician in some limited area of application. In a society of these technicians working together in various fields then we have another level of collective pseudo or virtual intelligences operating together and where my role is in that of course is to utilize them and input my creativity which will be very hard to duplicate in machines. We have two eyes, two ears and that's all we really can hear and see and we basically have one line of language in our mind. We can't listen to six or seven conversations at once. This machine would, could and would handle just tons of input. You will have a very capable companion not capable of certain sense but we'll have a much bigger memory that you ever have and recall. We'll can be able to make decisions much faster than you can using millions of facts instead of four or five that you can juggle in your head. They may way exceed our ability to assimilate and digest facts. But there is really not even a seed of creative intelligence in that. Different thinking machines will actually have different personalities in a sense. The idea that they're a servant that execute just what you've defined that era is going to end and we're going to be in a whole different arena where they're going to be operating with the human as a partner. I think some of the biggest problems in computing are going to be social and political rather than technical. The more stuff about yourself that you put online the more the less control you have over it the more people are going to have access to it that maybe you don't want to have access to. As fast as the society as we have today I think it's only going to get faster. I think the biggest problem that computers can cause is the lack of personal contact and the human touch. Sometimes the fascination with the power of the tool steers you away from actually applying it to solve human problems. I also have a fear of losing craftsmanship and skills. I think that as we have a computer capability that is going to be able to handle more and more of our routine tasks that one of the incredible challenges we face is to educate kids how to solve problems how to be creative how to be innovative as opposed to regurgitating facts and figures because that's what computers are going to be able to do best for us. We'll probably have social programs to make sure that we get cellular computers to everyone so they can have a say in the way the world operates. You know we talked today about the potential rift between the haves and haves not growing wider. I kind of turned that into the future concern for the nose and the nose knots. It's one of those equalizer tools I think that can be useful to a lot of people to help them overcome some of the disadvantages they might have as an individual or small group. I'm interested mostly in adaptive technology since it applies to me both in my professional and personal life. My hopes maybe like I said in my professional life is to have things that I can use in the office so that I can identify much more than I do now as far as documents and things like that and some added additions hopefully being able to be compatible with graphical user interface using speech or maybe even using voice recognition systems and for my personal life maybe a computerized car. I have a lot more dynamism, a lot more mobility, a lot more adaptability, a lot more independence, a lot more requirement for responsibility, for education, for initiative. While we have all of these things they have not brought the world happiness. It's brought it closer together made a smaller world. We seem not to understand one another that seems to be quite prevalent. Computing can only be a tool that's going to help bringing information to us. I think that the thing that humans are probably for the next 20 or 30 years going to always be more capable than computers is of imagination of conceptualizing and visualizing things and asking the question of how could this be different or why is this and then posing that question the computer and maybe finding an answer. More and more I think what I see is that what knowledgeable people are capable of imagining they're capable of doing.