Special thanks to Andy Kirk for answering 6 questions about his recently featured book – Data Visualization: A Successful Design Process
Andy Kirk is a UK-based freelance data visualisation design consultant, trainer, author, speaker and editor of visualisingdata.com. He is founder of Visualising Data Ltd, undertaking design consultancy and delivering training. After graduating from Lancaster University in 1999 with a B.Sc (hons) in Operational Research, he held a number of business analysis and information management positions at some of the largest organisations in the UK. At the start of 2007 Andy had his career-changing ‘eureka’ moment when he discovered the subject of data visualisation and unlocked a passion that he has pursued ever since. After undertaking a Masters by Research M.A (With Distinction) programme at the University of Leeds, in February 2010 Andy launched visualisingdata.com with the mission of providing readers with inspiring insights into the contemporary techniques, resources, applications and best practices in this exciting subject. His consultancy work and training courses extend this ambition, helping organisations of all shapes, sizes and domains master the analysis and communication of their data. – From Andy’s Homepage
#1 – Visualising Data feels like such a new thing to the internet. Can you give us a brief history of the field?
Data visualisation is likely to be a new term to many people: it is one of those subjects that you don’t know about until you know about it, if that makes sense? Yet, it has been around in some form for over 250 years and much longer if you broaden the definition/discussion to mapping and information graphics. A rapid run through the history of the field must start in the late 1700s with an acknowledgement of the role of key figures such as Joseph Priestley, William Playfair and Charles Joseph Minard. These were amongst the early pioneers in developing techniques for visually representing data. They conceived methods such as timelines, bar charts, pie charts and line charts: approaches that are still commonplace in contemporary forms of data communication. Things moved along steadily over the next 200 years or so, with the field very much the preserve of the specialist – typically statisticians, economists or engineers. It was in the 1960s that we saw some of the most influential attempts to study graphical methods and visual perception from a theoretical perspective. This generation brought us important foundation principles from the likes of John W Tukey and Jacques Bertin. The significance of this era cannot be underestimated, with an increasing body of research emerging and early technological developments offering new capabilities for the analysis and presentation of data.
Over the subsequent two decades many of the fundamental guidelines and techniques that shape the discipline today became established and influential figures such as Edward Tufte emerged with fresh perspectives that brought a new level of convergence between design and statistics. As we enter the 2000s the incredible progress of technology becomes the key catalyst. Firstly, through the exponential increases in the volume of recorded data and ongoing improvements in the widespread provision of access to it. Secondly, via the advancement of tools that provide powerful means for analysing and communicating this data, the ubiquity of which helping to fuel modern society’s demand for information. Thirdly, the Internet: the essential platform for creating and connecting a contemporary community of practitioners. Moreover, it provides a rich creative canvas for innovative new techniques to flourish, which leads us to the modern day and a field that is rapidly increasing in consumer popularity and creative participation.
For reference, there is a great in-depth timeline providing an overview of the key milestones in the history of data visualization http://www.datavis.ca/milestones/.
#2 – How did you yourself get into this field?
I am fortunate to have experienced a very tangible moment of discovery with data visualisation. It is something I wrote about it in an article a couple of years ago (http://www.visualisingdata.com/index.php/2011/10/open-thread-when-did-you-discover-data-visualisation/). Going back to my days at school, I was somewhat unique in my enjoyment of both art and maths. My intended career path would have been architecture were it not for my utter failure in grasping physics! Instead, my career followed a path of working in or around information and performance management. I have always held positions with a responsibility for analysing and presenting data so visualisation, as I know it now, was always something I did. However, I pursued this activity without any conscious or informed approach: it was all based on what I thought looked nice (or what my bosses liked, which meant it was generally horrible). I never really thought much deeper about it. At the start of 2007, when I working at the University of Leeds, I was given a graphical task that needed a little bit of online research. By chance, I happened to come across Stephen Few’s ‘Perceptual Edge’ website (http://www.perceptualedge.com/). Here, for the first time, I found a community discussing graphs and charts, talking about the rights and wrongs of making them, and using the term ‘data visualisation’.
The light bulb went on above my head, big time! I had discovered a new subject that perfectly harmonised my keen interests and experiences. It gave me something that I immediately knew I would want to immerse myself in and embark on a process of learning. At that time, I had an opportunity to undertake a Masters research programme and fortunately I was able to tear-up my original proposals and shape my study entirely around data visualisation. This study cemented my passion for the subject, so much so that I decided to continue learning, practicing and partaking in the field by launching my blog www.visualisingdata.com in February 2010. Through writing about the subject I found I was able to develop a more critical eye and refine my convictions. It also gave me a chance to connect up with other professionals and talented practitioners in the field. Over time, I was fortunate to secure a few opportunities to deliver training and provide consultancy to organisations that had discovered my site. This was on top of my full-time day job so as the demand grew I had a big decision to make! Ultimately, the choice was clear and I became a freelancer in October 2011.
#3 – Who are you hoping this book will appeal to? And why?
There are so many different characteristics and types of people interested in visualisation so I hope it would appeal to a range of potential readers. Irrespective of whether you are an experienced visualizer or a rookie just starting out, I would hope this book provides useful and practical guidance to anyone looking to learn more about data visualisation design and seeking to optimise their techniques. The intention is for it to be something for everyone with an interest in charting and graphical methods. For instance, you might be coming into data visualisation as a designer but need to bolster your data skills. You might be strong analytically but are seeking some inspiration on the design choices. You might have a great sense of ‘story’ but don’t quite possess the means for portraying this visually. It is important to define what it isn’t. If readers are looking for deep discourse in the subject’s theory, instructive software tutorials, or coffee-table glossies rich in beautiful imagery galleries, there are other books out there that offer this. No book can be everything to everyone. However, if a reader is seeking to focus on improving the effectiveness and efficiency of their design process and the critical thinking that surrounds this, maybe this will be a valuable read.
#4 – In your book, what essential tips do you give to people who want to visualise data?
I would say the primary focus of my guidance in this book concerns the organisation of one’s thinking – a process driven approach to visualising data. That is not to say it has to be a rigid “Part A goes in to Slot B” sort of mechanistic approach but, regardless of the size and complexity of a graphical challenge, there are certain sequences of thinking that will always significantly aid your approach. For many, the understandable temptation is to dive straight in to design work – the sexy and fun tasks – choosing a chart type or graphical construct that we really want to use and shoe-horning the data into it.
However, the thinking and preparation that goes before this design stage is critical and often neglected. Contextual factors such as carefully considering what you are trying to accomplish, who is it for, what data have you got – the raw material – and what insights it holds will have a huge influence over your eventual design choices. Indeed, it will save you so much more time and effort. There are many other fundamental tips contained such as the need to strive for a balance between form and function that suits your context, the emphasis on clarity over simplicity and the need to be aware of how and when visualisations can be incorrectly constructed, leading to misleading interpretation.
#5 – To you, what are the ingredients of a good data visualisation? And a bad one?
Without wishing to sound overly ‘Donald Rumsfeld’-esque (http://en.wikipedia.org/wiki/There_are_known_knowns), possibly for the second time in this article, one way to summarise what makes a good visualisation is that it should make the invisible, visible and the visible, invisible.
Making the invisible, visible concerns the patterns and findings contained within data. To the naked eye, when data is presented in its raw state, one is unable to readily perceive such insights. What good visualisation does is make these visible, enabling us to ‘see’ the data rather than just ‘look’ at it. A visualisation should add value to the portrayal of data, giving us a visual perspective that reveals something about the physical and quantifiable aspects of a given subject.
Making the visible, invisible concerns the design choices and execution. Good visualisation avoids creating barriers, causing delays or invoking confusion in the process of drawing interpretations. Readers should not be required to ask questions like ‘what does that colour represent’, ‘why is that positioned there’, ‘how do I zoom into that area’. The design choices should be focused on efficiently explaining or intuitively facilitating the process of consuming a chart or graphic. We shouldn’t need to overly think. We’re not great at that.
#6 – Are you working on another project/book that you can tell us about?
It is still in the very early stages but I am indeed in the proposal stages for a second book. I feel I have so much more to discuss about the subject and, together with my publisher, I am exploring different options and formats to achieve this, with some really interesting innovative digital ideas in prospect. Unfortunately, I can’t say much more about it at this stage but I’m really keen to get going with it during 2014!
[Image Credit: http://grubstreet.co.za/wp-content/uploads/2012/12/AndyKirk.jpg ]