But while traditional engineering has usually ended up finding ways to avoid searches for the limited kinds of systems it considers, the phenomenon of computational irreducibility makes it inevitable that if one considers all possible simple programs then finding particular forms of behavior can require doing searches that involve irreducibly large amounts of computational work.

And in a sense this means that if one tries directly to produce specific pieces of technology one can potentially always get stuck. So in practice a better approach will often be in effect just to do basic science—and much as I have done in this book to try to build up a body of abstract knowledge about how all sorts of simple programs behave.

In chemistry for example one might start by studying the basic science of how all sorts of different substances behave. But having developed a library of results one is then in a position to pick out substances that might be relevant for a specific technological purpose.

And I believe much the same will happen with simple programs. Indeed, in my experience it is remarkable just how often even elementary cellular automata like rule 90 and rule 30 can be applied in one way or another to technological situations.

In general one can think of technology as trying to take systems that exist in nature or elsewhere and harness them to achieve human purposes. But history suggests that it is often difficult even to imagine a purpose without having seen at least something that achieves it.

And indeed a vast quantity of current technology is in the end based on trying to set up our own systems to emulate features that we have noticed exist in ordinary biological or physical systems.

But inevitably we tend to notice only those features that somehow fit into the whole conceptual framework we use. And insofar as that framework is based even implicitly on traditional science it will tend to miss much of what I have discussed in this book.

So in the decades to come, when the science in this book has been absorbed, it is my expectation that it will not only suggest many new ways to achieve existing technological purposes but will also suggest many new purposes that technology can address.