28 of 31 people found the following review helpful:
5.0 out of 5 stars
Clear thinking on how technology evolves and shapes the economy, September 7, 2009
This review is from: The Nature of Technology: What It Is and How It Evolves (Hardcover)
An engaging and thought provoking book, Arthur provides a powerful framework for understanding how technologies evolve and are a key driver of productivity growth. According to Arthur (and he does a good job of demonstrating his case), technologies are based on interactions with natural phenomena that are composed into modular systems of components that grow into domains with their own conceptual languages. Because the systems are modular, they can leverage the combinatorial explosion and once a certain technology reaches a critical mass of components and interfaces it can evolve rapidly, entering new domains and exposing new natural phenomena to interact with. Arthur provides many examples that are interesting in their own right - from the evolution of airplanes and turbojets to genetics and even gearing systems or sorting algorithms.
One test of a book is if it draws you towards additional reading that you might not have otherwise discovered. Arthur's book caused me to run out (to Amazon) and order Colum Gilfillan's 1935 book Inventing the Ship and decide to finally read Donald McKenzie's book Knowing Machines. Thank you.
I do have a few quibbles. I think Arthur makes a serious conceptual error in making natural phenomena the `genes' of his system. I understand the temptation, but I think the metaphor is based on a misunderstanding of how genes actually function in living systems (see for example Lenny Moss' book What Genes Can't Do). The primitive elements in technology evolution can not be natural phenomena themselves but how humans (and other species) interact with these phenomena. I am not sure how to formalize this, probably something like a `theory in use" of cause and effect for natural phenomena, not something as formal as a scientific theory, more the rules of thumb and satisficing that we use as we interact with our world.
There are also some conceptual frameworks that could be used to complement Arthur's approach. I think the most important of these is that of design spaces, and the idea that technological progress is based on the expansion of and improved search over design spaces. For me, Stuart Kaufmann's work is foundational here. Other work that complements Arthur's is Baldwin and Clark's wonderful book Design Rules (I hope that Volume 2 actually comes out one day) and the many applications of design patterns that are spreading from Christopher Alexander to the software industry to many other areas of endeavor. I personally find work in mereology useful in thinking about part-whole relationships and in converting combinatorial explosions into navigable design spaces, see for example Roberto Casati and Achille Varzi on Parts and Places.
Arthur's approach is going to need some formalization and a lot more application, but I think it proposes a useful way forward. It will be interesting to see how these ideas are applied to technologies such as markets and financial instruments, as well as new designs for organizations such as the fourth sector.
Help other customers find the most helpful reviews
Was this review helpful to you? Yes
No
22 of 24 people found the following review helpful:
4.0 out of 5 stars
A coherent theory of the development of technology, November 12, 2009
This review is from: The Nature of Technology: What It Is and How It Evolves (Hardcover)
W. Brian Arthur, who is both an engineer and an economist, has thought a lot about the logic of technology. The strength of this book resides in how he pulls his observations together into a clear and coherent theory of how technology evolves. Arthur repeats himself to some degree throughout (one could read just the preface and the last chapter to grasp the main elements of his theory), but the prose is relatively jargon-free and straight-forward.
All technologies, as Arthur defines them, (1) entail a means to fulfill a human purpose and (2) involve an assemblage of practices and components (both devices and methods). "Technology" can also mean the entire collection of devices and engineering practices available to a culture.
The essence of technology, Arthur suggests, is a phenomenon or set of phenomena captured and put to use, a programming of one or more of "truisms of nature" to our purposes (for example, burning certain fuels produces energy we can employ in many ways). The history of technology, he proposes, is one of capturing finer and finer phenomena, enabled by earlier technology.
As he sees it, technology provides a "vocabulary" of elements that can be put together in endlessly new ways for novel purposes. Technology is "autopoietic," or self-creating, Arthur believes. It creates new opportunity niches and new problems, which call forth still more new technology. The economy is in a state of perpetual novelty, unsatisfied, roiling constantly.
According to Arthur, technologies often group together into "domains" based on the natural effects they exploit. He believes that, "A change in domain is the main way in which technology progresses" (for example, a shift from mechanical to electronic controls, or from analogue to digital electronics).
Just because we have a theory for how technology evolves does not mean, however, that we can accurately predict the technological future. There are many indeterminacies, Arthur says. He recognizes that the investment and publicity environments, for example, matter in determining what gets developed and adopted, and at what speed, but he doesn't say much about these matters.
Yet if technology has a logic of its own, why does it proceed at a different pace and on a different course in different places? The obvious answer is, I believe, that culture matters too, in all its manifestations (business systems, religious beliefs, governance structures, and so on). To be fair, Arthur says he made a deliberate choice to focus on the logic of technical creation (and not on the people or institutions who do it), and he treats societal institutions themselves as technologies, but as a consequence he sometimes comes across as too techno-centric.
While Arthur does an admirable job of presenting historical examples (drawn mostly from the past two centuries), he has been selective, naturally latching on to cases that support his contentions. Do not expect a broad history of technology in the sense of a systematic survey of a wide range of developments in any given historical era. Thus we don't know for sure from this volume alone how well his theory might hold up against a more inclusive consideration of historical developments, especially across cultures.
Because Arthur's concept of technology is so broad (pretty much anything that fulfills a human purpose counts), it raises several boundary issues; for example, where should one draw the line between science and technology? He concedes that it would be stretching things to call Newton's explanations, for instance, "technologies" and proposes that it is better to think of scientific explanations as purposed systems that are "cousins" to technology.
In the end, though, such fuzziness may not be much of a detriment, because Arthur's broad conceptions lead him to provocative insights. For example, he rejects the idea that technology is simply the application of science and he observes that many technologies came into being without drawing on science directly at all (for example, powered flight). It was only when the phenomena driving technology began to fall below the threshold of unaided human observation (such as electrical and chemical phenomena) that science began to play more of a role, he proposes.
Arthur also has engaging things to say about similarities and differences between technology and biology, about how engineers work, about how economic "needs" are generated, about our conceptions of nature versus technology, and about several other related subjects that should be of interest to many general readers.
Help other customers find the most helpful reviews
Was this review helpful to you? Yes
No
25 of 29 people found the following review helpful:
5.0 out of 5 stars
Powerful thinking, September 10, 2009
This review is from: The Nature of Technology: What It Is and How It Evolves (Hardcover)
Books like this are published rarely -- maybe once every 10 years. Brian Arthur has done a masterful job of presenting new ideas about technological evolution and innovation in a way that is engaging and accessible. The Nature of Technology is beautifully written. That's a recommendation in itself, but it is the new thinking that is most significant. Arthur explains how each of our technologies is a system, assembled from other technologies... ad infinitum. Every component provides an essential function in support of the whole. As components improve, or new components are substituted with enhanced functionality, the system evolves. Our technologies are now deep and complex, with many nested levels.
Arthur's model nicely explains accelerating change. In a simpler pre-industrial world, we had fewer things to combine. Today we have a seemingly infinite number of technologies to work with, and can combine them in an infinite number of ways. Add a new technology and the combinations multiply. One reflects on how quickly the Internet has been embedded in other technologies in ways that have created widespread systemic change.
Technology, Arthur says, harnesses phenomena to deliver its functionality. We can see this in the evolution of computers, where calculating machines were first based on mechanics, later computers harnessed the forces of electricity and magnetism, and researchers today grapple with the challenge of creating a computer based on the counter-intuitive laws of quantum physics. This dream has not yet been realized, but it illustrates Arthur's principle. Scientists and engineers are working on multiple fronts to transform ethereal quantum phenomena into a reliable and concrete computational machine.
Arthur's framework leads in some interesting new directions. While computers use natural phenomena to perform their function, they create new phenomena -- in the form of information -- that can be used in other ways. Emergent phenomena created by our technologies are fertile ground for still further innovation.
It's a rare book that presents new ideas on every page. This is one of them. The result is an important new framework for thinking about technology and how it evolves.
Help other customers find the most helpful reviews
Was this review helpful to you? Yes
No
