Book Reviews

Börner, Katy. Atlas of forecasts. Modelling and mapping desirable futures. Cambridge, MA: MIT Press, 2021. xi, 214 pp. ISBN 978-0-262-04595-7. $39.95

This is the third "atlas" produced by the author (in association, it must be said, with many others, who are acknowledged in an extensive References and credits section, and in the Acknowledgements). Her Atlas of science was awarded ASIST's information science book of the year in 2011, and in 2015 she published the Atlas of knowledge. This atlas, is very similar in style and structure to the previous books. Like the others, it is more of an exhibition catalogue than a standard reference work, although there is extensive treatment of many different kinds of forecasting models and similar tools.

Reviewing a work of such complexity is not an easy task: there is so much to take in and think about. We can begin, with the author's statement of the aims of the book:

This Atlas of Forecasts introduces major model classes and applications to a general audience, using real-world examples and advanced data visualizations. It demonstrates the power of models to understand and predict the structure and dynamics of diverse sociotechnological systems, from micro to macro levels.

To achieve this aim the book is divided into five parts: Part 1 is an introduction and brief history of models. I was a little surprised at how recently the author dates the emergence of models. She does mention the Delphi Oracle as a kind of model for the Delphi method of forecasting or problem solving, and John Graunt's 1662 publication of his analysis of the Bills of Mortality, the first book of public health data, but I would have thought that conceptual models, at least, must go back Plato.

Model types are the subject of the second part of the book. Here we find page spreads on topics such as predictive models, probability theory, control theory, epidemic models, and network models. Each spread has appropriate illustrations, citations to other points in the book and previous "atlases". Further references are supplied in the section at the end of the book.

The sub-title of the book refers to "desirable futures" and in Part 3 the factors (here called "substrates") that affect our possible futures are described and illustrated in relation to eduction, science, technology and policy. The substrates are population health and education, natural resources, climate and weather, transportation infrastructure, digitization, and urbanization. Models related to each or these are explored on the following pages, divided into micro (individual), meso (team, institution, regional) and macro (population, global) levels. For example, a model of learning and forgetting in corporate production systems is presented at the micro level in the technology domain, while at the macro level in policy a diagram shows various global risk factors, such as climate change and wealth inequality.

The author's books have an associated exhibition and Part 4 shows the maps that have been added to the exhibition since 2012. The range is very wide, from William Newman's Geologic time spiral, through John Nelson's map of the location and intensity of tropical storms since 1851, to Martin Vargic's Map of the Internet (Note that the link leads to a more recent map than that shown in the book). One can spend an enormous amount of time poring over these maps!

Part 5 discusses the various opportunities for modelling in relation to considering desirable future states. It includes modelling opportunities, reducing human bias, managing risks, and creating actionable forecasts.

The list of references and credits has already been mentioned and there is, finally, an extensive index.

The amount of work involved in producing a work of this kind is obviously enormous and one understands the reason for the time gap between the production of one "atlas"e; and another. One can say, however, that they are well worth waiting for!

Professor T.D. Wilson
Editor-in-Chief
August 2021

How to cite this review

Information Research is published four times a year by the University of Borås, Allégatan 1, 501 90 Borås, Sweden.