Learning the new kind of science

There will, I hope, be many who want to learn about what is in this book, whether out of general intellectual interest, to apply it in some way or another, or to participate in its further development. But regardless of the purpose, the best first step will certainly be to read as much of this book as possible with care. In time there will doubtless also be all sorts of additional material and educational options available. But ultimately the key to a real understanding is to experience ideas for oneself. And for the new kind of science in this book this is in a sense unprecedentedly easy, for all it requires is a standard computer on which to do computer experiments.

At first the best thing is probably just to repeat some of the experiments I describe in this book—using the software and resources described at the website, or perhaps just by typing in some of the programs in these notes. And even if one can already see the result of an experiment in a picture in this book, it has been my consistent observation that one internalizes results of experiments much better if one gets them by running a program oneself than if one just sees them printed in a book. To get a deeper understanding, however, one invariably needs to try formulating experiments for oneself. One might wonder, for example, what would happen if some particular system were run for more steps than I show in this book. Would the system go on doing what one sees in the book, or might it start doing something quite different? With the appropriate setup, one can immediately run a program to find out. Often one will have some kind of guess about what the answer should be. At first—if my own experience is a guide—this guess will quite often be wrong. But gradually, after seeing what happens in enough cases, one will begin to develop a correct and robust new intuition. Realistically this seems to take several months even for the most talented and open-minded people. But as the new intuition matures, ideas in this book like the Principle of Computational Equivalence, that may at first seem hard to believe, will slowly come to seem almost obvious.

For someone to assimilate all of the new kind of science I describe in this book will take a very significant time. Indeed, in a traditional educational setting I expect that it will require an investment of years comparable to learning an area like physics. How long it will take a given individual to get to the point of being able to do something specific with the new kind of science in this book will depend greatly on their background and particular goals. But in almost any case a crucial practical step—if it has not already been taken—will be to learn well Mathematica and the language it embodies. For although most simple programs can be implemented in almost any computational environment, not using the capabilities of Mathematica will be an immediate handicap—which, for example, would certainly have prevented me from discovering the vast majority of what is now in this book.