Overcomplicated

June 20, 2016
Arbesman (The Half-Life of Facts), a self-described "complexity scientist," presents a new framework for understanding and working with complex technological systems in this thought-provoking treatise. Arbesman argues that technological systems have become so complicated that not even those who design them fully understand how they work, nor do they always know what to do when their systems fail or return unexpected, possibly catastrophic results. He illustrates this through numerous examples of flaws or breaks in increasingly sophisticated systems such as traffic control, the stock market, machine translation, and medical devices. Despite the damage caused by bugs in programs and other system defects, they are valuable in that they reveal components of the technology at work. Arbesman proposes that programmers and designers approach technological complexity with the mindset of naturalists, who study the natural world by cataloging its variety. By treating failures or glitches in the system as by-products of the evolution of complexity, he says, we can predict paths to new technology and better prepare for the unintentional consequences. Arbesman moves confidently through his layered argument in smooth, clear prose. He convincingly conveys the relevancy of the subject to all types of industries, but his book will mostly appeal to readers interested in the theoretical component of computer science.

June 15, 2016

Every computer user has experienced programs that inexplicably crash or, less frequently, exhibit unexpected capabilities. Such behavior arises because the system has reached a level of complexity that is beyond even its creators' skills to disentangle. It may seem strange that human beings can produce such entities. However, Arbesman (senior adjunct fellow, Silicon Flatirons Ctr. for Law, Technology, & Entrepreneurship, Univ. of Colorado; The Half-Life of Facts) points out that this situation is not unique to computer software but can also be found in legal and tax codes and in our infrastructure. In particular, the author stresses the similarity between such systems and biological evolution, in which the process of accretion adds new functions to the original system, imperfectly, but fitting just well enough to make them work. Unlike many who fear the singularity or who are overwhelmed by technological incomprehensibility, Arbesman offers a more optimistic view. He asks readers to accept that at best we can grasp only an overview of the situation, to be content with understanding the portions that are accessible, and to rejoice in being members of a species that has the intellect to produce such marvelous creations. VERDICT This is a short, comprehensible book that should find an audience.--Harold D. Shane, Mathematics Emeritus, Baruch Coll. Lib., CUNY

Copyright 2016 Library Journal, LLC Used with permission.