Last night I attended my first “Big Data Debate” (@BigDataDebate) event. These are a series of regular events focusing on all things data. Big Data in Media, Finance, Data privacy, open source data etc. http://www.meetup.com/BigData-Debate/
First up was Andy. His presentation focused on 4 pointers of why we visualise information. He spoke about pre-attentive attributes and Gestalt Principles of visual perception as his first pointer, moving on to critique David McCandless visualisation of colours as emotions as an example of why you should engage with function, not candy. He contrasted this with Simon Scott’s visualisation of the Iraq death toll.
Andy’s third pointer was to focus on the message of your visualisation. To illustrate this he completely changed the feel of the Iraq viz by simply inverting it, changing the title and the colour. Very impressive.
Finally Andy highlighted the need to explore and discover using your data. To illustrate this he broke out Tableau and did a live demo focusing on his Spotify data. See this blog post for more info on that.
Then it was time for a panel debate / discussion featuring the following. From left to right..
Andy Cotgreave: Data visualisation expert and Tableau evangelist, Tableau
Paul Joyce: CEO, Geckoboard
Jim Hodgkins: Managing Director of Marketing Services, Visual DNA
Richie Barter: CEO, AltViz
Got some very interesting topics of discussion, including the following
- What is it about dashboards that make them a good approach for data viz?
- How do you ensure your top line allies with that of the consumer when designing vizzes?
- Critique of bling vs function
- At what point do you intervene if your customer disagrees with your interpretation of best practices?
- Is it part of a data viz professional’s job to educate their audience?
- Is there a role for animation and motion graphics
- What do you see as the next step change in data visualisation technology / principles?
A wide variety of answers to these questions from the panel including discussions on how dashboards have been made to do too much by many software companies, the importance of educating users and maintaining best practice and how bling charts may be ineffective but can be very useful at bringing in new users to the world of data visualisation. Some of the new tech ideas focused around mobile/wearables/quantified self and also the use of story telling in data visualisation.
Unfortunately there wasn’t much time for audience Q&A, and that wasn’t helped by some dodgy microphones. I was hoping that part of the session would be much more dynamic.
Overall well worth attending. Connected with a few Twitterati for the first time and also had a very good post session chat with a couple of the panel members. I’m particularly interested in the work of AltViz.co – keep an eye on those guys.
Looking forward to the next one. Keep an eye on http://www.meetup.com/BigData-Debate/