Steven Levy is the Senior Editor and chief technology writer for Newsweek and contributor to lots of other places, including wired. His book A-Life focuses on the topic of artificial life and how its theories came about. From reading this book I have concluded that Levy is an evolutionist. He always mentioned that about the person he was writing on and it seemed in an agreeing manner. Levy said at one point, "something equally amazing apparently occurred many years ago on this planet when simple molecules somehow developed into a complex system that eventually bore what we now call life."
Artificial life is the realm of computer science that deals with using computers to emulate real life, such as human thought, reasoning, reproduction, colonization, mobilization, and other areas. I came into this book fairly interested in the subject matter and was excited to see what A-Life, artificial life, was going to be bringing in the near future. My perspective on this book was totally wrong. Levy outlines the dealings of a-life since the beginning of the previous century. Each chapter focuses on a person who devoted themselves to a particular field in a-life and those who helped them. The following are a few examples of what Levy had to say.
John von Neumann, the great mind and pioneer in understanding the computer as a logic machine. Neumann postulated The General and Logical Theory of Automata, or self-operating machines. He said, through the understanding of automata we can understand life. At a lecture he gave in Pasadena, CA at the Hixon Symposium Neumann's largest question "dealt with the concept of self-reproduction. Could an artificial machine produce a copy of itself that would in turn be capable of creating more copies?" His ideas and lectures distinguished him as the father or artificial life.
The quest for making artificial life started with the scientists redefining the definition of life. Previously life was "dependant on a certain quality bestowed on its parts." The author of the book describes this in terms of the scientists who saw this idea as absurd as "a trespass on the divine" it was God who breathed life into humankind (Genesis). Aristotle "believed that what distinguished organisms from their inanimate surroundings was possession of a soul." With the introduction of Newtonian ideas and the industrial revolution life was viewed by more to be "a mechanistic process." Finally the basis of life was considered to be information. Information steeped in a dynamical system complex enough to reproduce and to beat offspring more complex that the parent.
The author then asked, what about one of the largest differences in life compared to a-life, that of a-life consisting of man made screws and bolts to that of real life consisting of organs and bone? The maker of the Turing Machine and Alonzo Church presented the Physical Church-Turing Hypothesis which claimed to duplicate the function of nature. A-life became a new term, focusing not on artificial intelligence or the mind, but on self-reproduction or life itself. Many, from the claim of Turing and Church, believed that life could be emulated by automation, thus life being automata. von Neumann's tape was like that of the later discovered DNA strand. He made the important observation that life was not only grounded in information but also in complexity.
Self-Replicating Systems was a topic that Levy talked upon a few times in his text. This is the idea of making robots that could be self contained and have no need of human help after its beginning. They could be used to prepare to way for humans to colonize Mars or the moon. It was a fun idea, but many moral issues arose and the idea became just a thought. People worried about these systems of robots taking over and not caring that we created them for a purpose.
The use of Cellular Automata became huge when John Horton Conway introduced it in a great new way. His version of CA's (cellular automata) is still well known, it's the "Game of Life." This is the game that our professor Levy showed us in CS111. This was once a huge area being looked at by early scientists of the previous century. The first games of life were done by people using different colored stones and changing and pausing between each time state. Eventually with the help of R. William Gosper at MIT, Conway was able to develop and, or and not-gates in Life and show that von-Neumann's ideas could be understood in much simpler ways. Many people who used CA's thought that a sufficiently large CA's could replicate human thought. They also mentioned that in theory a CA that replicated the thought of a small animal would be the size of our galaxy and a CA representing a single cell would be the size of our solar system. Not very practical, huh?
Another man Stephen Wolfram made 1CA's, or one dimensional cellular automaton, and from them showed how a complex look came from very simple rules. Maybe life was more like this, simple producing complex. Other aspects of real life situations were looked into as time went on. A man named Craig Reynolds simulated the way that birds flocked together.
James Doyne Farmer wanted to find rules or laws that showed how entropy worked. Also proving or disproving the veracity of evolution. Both of these proofs were not done just to prove them, but to understand greater ideas. He wanted something that would provide the laws to what happened in weather, economy, or biological organisms. He and others during the "Garage-Band Science" years searched on how to make artificial biology. Langton, a scientist during this time invented an informational cell that was capable of replicating itself. At this time the experiment showed that evolution and biological life could be shown through information. Out of his work and in collaboration with others, people began to ask the question, "What is Life?" a question that occurs over and over in this field of study.
Stuart Kauffman studied human cell differentiation. He began to ponder on how to study the genes in humans. While others wanted to study only segments of the 100,000 or so genes we have, he wanted to take the whole picture into account and work from there. He used the language FORTRAN to write a program that simulated 100 neural connections of the genes. From this he found that the state of the machine when ran would simulate a periodic attractor, or in the chaos theory, a looping cycle.
Stanley Miller conducted an experiment in which he produced amino acids from simulated lightning and other various elements. People saw this as proof that life did not just come from miraculous ways, but from something rather mathematically dictated. From this information Kauffman produced a grand new idea about the origins of life. This theory was different than others in that the possible occurrence of it was very likely, unlike other theories that are based on odds.
John Holland was a very bright man who at MIT studied under Norbert Wiener in the area of cybernetics. He began working with IBM and made a "lab rat" with them that could learn a maze and find it's way out. This was the beginning of AI. Holland was known for the idea of a parallel processor, for he thought for a computer to act like something in this world, it would have to be able to think many things at once. Later, after reading a book on genetics and evolution, "Holland was entranced" with the idea and began what would be his field for a very long time. He postulated what he called the Genetic Algorithm. The algorithm took strings of binary numbers and translated them into chromosomes for itself. An experiment with computer ants proved this algorithm's worth. The main thrust of the GA was that of evolutionary artificial life. And it worked, but not very many others took to the GA. The language LISP, written by Koza, was more effective than a GA and solved equations like the motion of planetary bodies, using pure evolutionary tactics.
There were many others talked about in this text with equal importance to those mentioned above including: William Hillis and the digital parasite, Aristad Lindenmayer and Przemyslaw Prusinkiewicz used the idea of embryological paths to a-life, Rodney Brooks and the mention of Walter and his tortoises, and finally the Strong Claim or computer viruses brought into greater light by Fred Cohen. In this book the basis for most of the ideas that proved the possibility of A-Life was that DNA and other fleshly things used states and conditions for making life. They do not seem to try to answer the question of whether or not this can be imitated in the digital world, but rather how do we imitate it. This is a very optimistic viewpoint considering that Dr. Brooks always kept in the equation that this type of calculation may not be possible in the silicon world. Most of this book is a history lesson in the area of artificial life. For someone that is interested in knowing the basics of A-Life and wanting to study it, this book is a great starting point. Outside of that desire of using the A-Life ideas in ones own work, this book is not really something that should come at the top of a "to read list", or even in the middle for that matter. But I do recognize that he meant this book to be a report on artificial life. In that light it was a good book.