Visualizing Data -- book review

[NewsBot]

For the Phantom Book Buyer -- thank you again for the book, here's a review.

 

Visualising Data, Ben Fry, O’Reilly

The first thing to note is that this book was not quite what I was expecting:* I was expecting a theoretical look at how to visualise data with lots of nice diagrams showing visualisations and techniques for visualisations.* That’s not what’s in the book.* The book is full of code! The books sub-title should have given me a better clue: ‘Exploring and Explaining Data with the Processing Environment’.* Note the last part ‘the Processing Environment’. The Processing Environment is actually a programming IDE and language based on Java (see http://processing.org/). Each chapter of the book consists of an extended example, talking through most of the code and adding extending the code as the chapter progresses. The book is actually based on the PhD thesis of the author, so I find it strange that is consists of a series of specific examples, with less information on the basic principles.

OK, so I’ve recovered from the unexpected contents of the book.* Onwards and upwards.

The first four chapters were something of a disappointment to me.* I have a science background, so I have done some data visualisation before and the basics covered here were nothing new (chapter three is basically how to draw a data plot over time).* However there were a couple of information gems buried in there amongst all the code.

Chapter 5 was where is started to get interesting for me.* The initial parts of the chapter concerning screen scraping from a website were nothing thrilling if you’ve ever parsed data before, but once this part was done came the interesting bit: representing data in a novel way and animating it to display the data over time.* Hooray, finally something to grasp my interest.

Chapter 6 – a scatter plot of US ZipCodes.* Yawn.

Chapter 7 – Ooh! Treemaps!* Interesting again. Unfortunately, as with many places in this book, the code of interest to me (to actually create the treemap) is imported as an existing library and the example serves to show how to use the library.

Chapter 8 – Graphs!* Everyone loves graphs! Don’t they?* Ok, the first part of the chapter deals with importing Java code into Processing.* Hmm.* Anyway, at least we’re drawing Graphs, perhaps we can cover some of the complicated options. At this point the book takes a turn to complexity and starts using a library called ‘anenome’ do draw some advanced graphs.* At this point, the book steps out of the Processing IDE and uses Eclipse.* This chapter feels a bit more like it. Proper programming, a proper task (analysing Apache Log files), and some data representation that I’d never seen before. Shame it was so short!

Chapters 9 and 10 deal with acquiring and parsing data.* I’ve you’ve ever done anything like this before, it’s interesting to see how someone else has done it.* However, at about 20% of the page count it felt out of place in a book on visualisation.

Overall I spent most of my time with this book thinking ‘I could do this in WPF! Perhaps I should start a C# library?’, *rather than being educated about data visualisation techniques.

Overall score: 6/10.

Summary:* too much example and not enough meat.* I would have liked to spend less time reading through specific code examples and more time on the underlying principles. If you’ve never done any data visualisation or data parsing, this book might be of use.

 

Herbie

 

 

More...

 

View All Our Microsoft Related Feeds