JOB: The evolution of the microprocessor over the past 25 years has gone beyond anybody's wildest expectations. At the first Bionomics conference in 1992, you remarked that you and your team at Intel could only envision 7 or 8 applications for the microprocessor! Today, the microprocessor is everywhere. We assume it was not modesty that prevented you from imagining the explosive growth in the microprocessor.

FF: Price/performance allows you to do only certain things at a point in time. The team at Intel that worked on the first microprocessor were pragmatic people. Early on, we saw applications for computation, table top calculators, cash registers, and banking terminals. Other applications we saw were controllers for gas pumps, elevators, traffic lights, and things like that. Some applications were developed more quickly than I thought. For example, the 4004 chip was used to control blood analysis instruments, and thus had medical applications early on. The 8008 chip came after 4004 and that broadened the range of applications in terminals and higher level calculators such as table top computers.

I do not believe that other members of the team at the time saw personal computers as a social phenomenon. Nor did I. At the heart of the birth of the microprocessor was enthusiasm. We were an enthusiastic group and our enthusiasm drove the development of the microprocessor. Today, the Internet is experiencing a similar evolution where enthusiasm among the new players in a field leads to new ideas and applications. But let us be frank: there is no certainty in looking at the future of the microprocessor or the Internet for that matter.

JOB: Looking back over the past 25 years, how do you feel about the evolution of the microprocessor?

FF: I feel good about the evolution of the microprocessor. I was fortunate to lead the team the produced the first Intel microprocessor, the 4004, and then the 8008, 8080, z80, z8, z8000. I feel fortunate to be part of this human movement. There is an element of fate in all this. Sheer energy and enthusiasm made this happen. Early on, the management at Intel did not really believe in microprocessors. The attitude was give it a shot and see if it flies. I was the one who was really trying to make things happen. At the time, the management at Intel was still thinking memories, and microprocessors were more of an afterthought. Even more than my technical accomplishments, I think the love and energy that I put into the project made the microprocessor happen.

JOB: At the last Bionomics conference, you gave the audience a glimpse of your current thinking. Could you tell us a little about the vision behind what you and your partner Carver Mead are currently working on at Synaptics?

FF: In essence, we are working on the creation of an autonomous, intelligent machine. We are interested in designing machines that can actually take care of themselves–machines that perform functions without being told what to do. These could be machines that perform functions such as cleaning floors. They would have sensory capabilities that tell the machine when the floor is dirty and needs to be cleaned. These would be machines that are happy leading a life of service! We envision these machines performing simple functions in an autonomous manner without being guided. Of course, these machines will need to have an enormous amount of intelligence–they need to have pattern recognition ability and they need to have sensory capabilities. The vision is to bring to bear a type of information processing that is sensory-based.

JOB: Does fuzzy logic come into play here?

FF: There is a lot of truth to fuzzy reasoning. But there is much more that we don't know than we do know. Our project and vision requires a deep understanding of how the brain works and how the brain can make sense of reality based on complex patterns. We have always taken the view that we tell the computer what to do. A sensory based machine would not need constant instructions from humans to function. Consider this: a computer can't write poetry. This is a complex task! But this is a task the brain does. We haven't even started to scratch the surface of the complexity and the capabilities of the human brain as it applies to sensory-based information processing.

JOB: In a sense, your work seems to bringing us full circle where computers become more human?

FF: I want to humbly understand how the brain does what it does and create machines that are inspired by this. In the process, that brings us closer to the way humans think and behave. There is something magical about humans, something special. The temptation to reduce human beings to a bunch of gates and flip-flops is really silly. I think that there is a dimension – a spiritual dimension to all of this. We will realize sooner or later that this dimension is unique.

We think it is possible to use the human brain as a model to allow us to build a machine that we can talk to and interact with. To be sure, to make a machine that takes on many human characteristics is a tall order, and something that I don't really want to fathom at this point. Frankly, I think it is premature to think of building a human machine. It is like the debate of philosophy 2000 years before quantum mechanics, only to find that the reality is even more crazy and complicated than anybody envisioned! We haven't even started thinking about modeling consciousness.

JOB: Let's move on to another issue. In your opinion, why does the U.S. command such a huge lead in high tech? Europe, where you and your family come from, appears to be far behind in creating the information technologies that will shape the future.

FF: I think there is a large cultural aspect to all of this. I believe it stems from what people value. People in Europe tend to value life and living comfortably more than creating something. In a way, they are less curious, and I believe this is related to culture. Religion has a lot to do with this. In Europe, there is still a lot of weight given to social status – where you come from, who your parents are, what school you attended. Europeans seem to be into categorizing people, putting bounds and limits on people. In other words, putting people in boxes. In the US, there is much less of this. The European attitude tends to stifle and limit people. In Europe, you are constrained by who you interact with on a daily basis. It is a cultural thing. I believe this applies to Japan as well.

JOB: We are curious as to why you are spending your time and effort on neural-net chips and not, for example, in creating Internet-related technologies. Any comment?

FF: The Internet is out of the hands of anybody. It is going to evolve in unpredictable ways. I enjoy working on projects where I feel I have certain amount of control over my destiny. Sometimes you win and sometimes you lose. The Internet is a rat race, a jungle.

JOB: Thank you Federico. See you at the next Bionomics Conference.

Related Links:

• Intel and The Smithsonian Institute present the 25th Anniversary of the Microprocessor
• The Microprocessor: A Biography
• HCS Virtual computer history museum
• Timeline of Microcomputers