Open thread: When did you discover data visualisation?

One of the questions that I am most interested in when I approach my Visualisation Insights articles concerns how people discovered the subject of data visualisation: when was it and what was the situation that led them to suddenly recognise or appreciate the existance of the subject?


The moment data visualisation appeared on my radar

I recall my moment vividly. About 5 years ago I was doing some (typically) last-minute analysis for a manager and he came up to my desk in a rush and said “Andy, can you do some spider graphs for that report because apparently they’re going down really well”. I had an idea of what he was talking about but thought I’d stick in a Google search and check I was on the right path.

I did the search, found a few references, realised I was probably thinking about the right thing and then decided to run a further search on the alternative, perhaps more common term of ‘radar graph’. One of the immediate items on the search result was a link to a pdf article entitled “Keep Radar Graphs Below the Radar – Far Below“, written by someone called Stephen Few.

I read the article and was captivated: for the first time, in an already well established analytical career, I had come across a critique of a graphical method (Stephen made it clear that I should be disobeying my orders!).

Excited, I explored the parent site, Perceptual Edge, in more depth and it soon became clear to me that here was a subject, data/information visualisation, that I had been completely oblivious to!

It was such a ‘Eureka’ moment for me because, for so long, I had been producing charts and graphs without any real sense of how I should be presenting them. I had always purely relied on my own taste and instinct. Now, I had stumbled across a subject field that was both completely new to me but equally fundamental to the career path I had followed. For somebody who loved art and maths and had studied Operational Research at degree level – a subject coined the “science of better” – I had a natural imperative to find out more.

I did, and the rest is history…


What’s your story?

So what was your moment, when you actually discovered data/information visualisation? I’d love to read your personal anecdotes in the comments section.


[...] Visualising Data » Blog Archive » Open thread: When did you discover data visualisation… [...]

Stephen MatthewsOctober 10th, 2011 at 10:36 pm

By background and training, I am a geographer. Now I teach geography as well as information technology. Geography, as an integrative discipline with a heavy emphasis on spatially-referenced data has always delved into data visualisation in its way. Not just maps, but graphs or various types, diagrams, cross-sections, flow charts, images (photographic, aerial/satellite), timelines, and so on, linked to place and space. Many times these are brought together in deliberate combination, both as sources for analysis or products of analysis. Many secondary school texts for geography moved to formats where a double page spread with minimal text and maximum visual data replaced pages of pure text a couple of decades ago. In my state of Australia, the end-of-secondary school geography examination, for well over two decades has been comprised of an A2-sized colour data sheet – which could easily grace the pages of this blog – that is full of sight unseen data for students to analyse.

So I pin my first conscious encounter with true data visualisations back into the early 80s as a student of geography at secondary school myself. Now as a teacher of geography and information technology, I find data visualisations integral to all that I do; still as sources and student products. In fact, just this year, the Year 11 (17 year olds), semester one state-mandated course for information technology is based largely on data visualisation techniques. To illustrate, from the course’s study design, students are to develop skills that demonstrate their ability to: “identify data required for particular purposes; select graphic representations appropriate to the purposes of creating solutions; select and apply appropriate design tools to represent solution designs, including the presentation of information; use spreadsheet software, and select and apply functions, formats, conventions, data validation and testing techniques to efficiently manipulate data and produce effective graphic representations; and explain why the solutions, including graphic representations, are suited to the purposes.”

Blogs featuring data visualisations such as this one have become an indispensable tool for teaching in both of my areas. The skills necessary to produce or interpret data visualisations are central to what I do. Whilst none of this is necessarily ‘new’, the growing availability of open datasets of all kinds, the increasing degree of use of data visualisations generally, and the use of the internet to disseminate it, have brought things to a new level. This shows no sign of slowing down either!

Ben JonesOctober 11th, 2011 at 12:02 am

Hi Andy – great question! I have a background in engineering, business, and Lean Sigma, so, like you, I became interested in understanding how to best use data to improve processes at work. I noticed like most of your readers that it made a huge difference how the data was presented. Often data was poorly visualized (er, manipulated) and people came to the wrong conclusions. I started researching graphical presentation of data and came across names like Tufte, Few and Tukey. I was hooked. Then I came across a growing online community of data visualization practitioners like yourself, Nathan Yau, and many others. I felt compelled to join the ranks and launched to your gracious welcome. This was more a gradual awakening than a “Eureka” moment, but highly transforming all the same.

Michel DekkerOctober 11th, 2011 at 7:16 am

Hi Andy, this question not only makes me think “when?”, but also “why?”. I must tell you the “why?” is easier to answer. I have been involved in reporting and analysis since 1995, and like you have created loads of tables and charts without really thinking about it effective presentation forms.
When I decided to start a new company around dashboards and reports in 2007 I was looking for ways to make these better (on a functional level). Then I came across Stephen Few’s work and started reading his publications and “Show me the numbers” (I believe his best book).
Since then I try to learn as much as I can and apply this knowledge as much as possible, and notice people appreciate this highly. But I am still surprised/disappointed most people involved in reporting and analysis (like me until 2007) have no clue about effective presentation forms. In most cases presentation is just the result of a personal preference, or because the tool used has the possibilities.
I am trying to find ways to help people to improve their data presentations, and blogs like these are a great help.

Andy KirkOctober 11th, 2011 at 8:24 am

Stephen, Ben, Michel – thanks so much for your fascinating perspectives, appreciate your time spent sharing your moments!

Stephen – do you happen to have a scanned/image version of the “end-of-secondary school geography examination A2-sized colour data sheet”?

Stephen MatthewsOctober 12th, 2011 at 9:26 pm

Hi Andy. Unfortunately, the data sheets (which have changed more recently to a less satisfactory, cost-driven data booklet) are copyrighted and reproduction is rigorously protected by the curriculum authority. The examination setters put a lot of time into their design and trying to ensure they picked a topic that was unlikely to have been encountered by students (one example being the global car production industry, combining maps, graphs, images, etc). In some ways it would be a disadvantage if students had done the topics in class with teachers, because there would be the tendency to form answers based on prior/external knowledge, when the skill is using and interpreting only the data which is presented in the examination. certainly the earlier data sheets were the best, and as time has gone on, they have become more formulaic and less original. Still, I thought the concept as it was originally conceived, as a true test of a student’s ability to deal with an unseen visualisation of data, was brilliant.

Andy KirkOctober 13th, 2011 at 8:09 pm

That’s a shame Stephen, but thanks anyway for sharing the tale – your enthusiasm for what it offered really comes across.