Author: Hunter Whitney
Publisher: Morgan Kaufmann
Pages: 320
ISBN: 978-0123877932
Audience: Anyone involved with analysing data and presenting results
Rating: 4
Reviewed by: Kay Ewbank

With the subtitle “New ways to visualize and make sense of data” what is in it for developers?

Data Insights is a book about how to make data more understandable. Hunter Whitney has specialized in designing user interfaces, and the premise of the book is that if you get the interface right, the user will be able to understand what’s going on more easily. As Whitney points out, there’s an amazing quantity of data that is theoretically available, but it’s all useless or misleading if we don’t know what it means.

With this in mind, Whitney has written a book that takes seven different looks at how you can view data. Whitney describes his approach as being a collection of ideas, observations, juxtapositions and conversations, and that’s a pretty accurate description.

The first chapter, titled ‘From Terabytes to Insights’, considers the overall problem that there’s an ever increasing flood of data to make sense of. He starts with a look at charts, the difference between data and metadata, and how to show complex data in simple ways. Put like that, this could be many database books, but the actual writing in the chapter is more like short stories – a day in his life referring to all the ways he interacts with data, for example.

The next chapter, titled ‘A More Beautiful Question’, is all about what questions you ought to be asking of your data, and more importantly, how to arrive at those questions. Chapter 3, Winning Combinations, looks at how different ‘visualization designs’ are suited to particular flavors of data. This is essentially ‘picking the right graph’, but there are some interesting case studies and discussions along the way.

Chapter 4 is titled ‘Pathways, Purposes and Points of View’. Whitney starts with the observation that some of the most interesting aspects of data are not individual data points but the pathways that connect them, and he goes on to look at ways to follow and analyze those pathways. Chapter 5, Views you can use, looks at how users interpret the data and graphs they’re presented with, and how ways to avoid them getting the wrong idea.

The next chapter, Thinking ... Machines, discusses the difference between what data mining lets computers do – search for patterns in large scale data – with what data visualization does – lets a person actually see the patterns in data. The question being asked is, when should you use a computer’s data mining software, and when should the human element be added. The last chapter, Hindsight, Foresight, and Insight considers the effect different computer technologies, such as clearer screens and touch screen devices, have already had, and takes a look at less common options such as 3D displays and holograms.

This isn’t an easy book to sum up. I really enjoyed reading it, and while doing so found myself agreeing with Whitney’s views and observations. However, once I’d finished it, I wasn’t sure what (if anything) I’d learned, or what another developer would learn. I think it’s worth reading to make you think more carefully about the way you display your data and results, the way you choose your data, and the way your apps interact with the user. I’d say the book is more about making you think about such things than any practical information about how to actually achieve the aim of better data visualization.