Data Visualization and Documentary Filmmaking
by Yoram Schaffer
It is not an article about functional analysis or molecular biology or science at all. It's about documentary filmmaking.
So what does the term Data Visualization have to do with documentary filmmaking and film art in general?
Data visualization is a general term which refers to graphical representations of data (usually statistical,geographical etc.) - i.e facts.
Documentary filmmaking is a broader term but most people will agree that it's about representing facts (reality) in an audio-visual, creative way.
Those simplified two definitions don't look very different from one another at first glance, but nevertheless, an attempt to discuss the use of data visualization in documentary filmmaking invokes fear and reservation among many filmmakers.
Before discussing those emotions and the possible reasons to them, let's present the exciting and promising domain of data visualization:
The above map represents population density in the Eastern part of USA. Dark colors represent low density and light - high density. The illustration is taken from Google Earth, where it's more dynamic and be combined with different layers of information. For example, if one wants to get a better understanding of why some areas are darker than the other, he can turn on other layers, which map parks and recreation areas. Naturally, no one lives there.
Why was data visualization used in the first place? - probably for educational reasons. I still remember large geography paper maps in elementary school which described different aspect each -- natural resources, population density, humidity, rain amounts per year etc. Back in the 1970's, no one called it Data Visualization but simply a 'map'.
There are plenty of other examples and data visualization is not limited to maps, but includes 3D speheres and other beautiful representations, which have their own aesthetic value. Actually, the Internet boasts thousands of data visualizations from artistic and scientific fields.
What discerns data visualization from other forms of art? - Basically, not even one pixel is there because the artist "felt like it" or had a divine inspiration to draw the wonderful bubble whic floats in space and changes its color while it turns. There is always a mathematical principle behind DV, since data is the monster that feeds the visualization. If there's no data - there's no representation. Data means raw facts, so even if you invented them as storytellers - these are still your own ad-hoc 'facts' for that matter.
But let's get closer to documentary filmmaking, or shall we call it storytelling? It might narrow the apparent gap between DV (Data Visualization) and DF (Documentary Filmmaking).
Let's turn to the work of Jonathan Harris:
Jonathan Harris combines between technology, statistics, data and storytelling. I hope he's not offended if I call him 'data artist'. The project whose picture appears above is Universe, which he made for Dailylife. Dailylife is a content service which gathers information from trusted publishers (mostly newspapers and bloggers) and organizers them in a smart way which looks like an online magazine - but with content coming from thousands of sources and not only the journalists which are regularly working for such magazines.
The Universe is the visualization of that data. A user enters a word which she wants to explore, relying on Dailylife database and an applet is invoked, presenting her with a cloud of stars, each one is a header for content (usually short text and image).
For example, I chose climate Change, one of the hottest (literally...) topics in Universe and Dailylife and a small universe opened up, containing starts, each representing a sub-topic concenrning an event connected to climate change from around the world - in this case Bolivian Qechuan natives sand outside the hall of the Conference on Vlimate Change and Mother Earth Rights which took place Tiquipaya, Bolivia during the week this articlw was written.
Jealously young and amazingly talented Harris already made a few data visualization projects, among them are Time Capsule which he made for Yahoo! and I Want You to Want Me which he made for the MoMa in collaboration with Sep Kamvar. In this project, data is gathered every few hours from dating sites and was represented on large screen as hearts floating in space - blue for men and pink for women. Men and defined by the way they describe themselves ('who I am') and what they expect to find in a spouse ('looking for'). On the screen, matchmakings are done in a graphical way, by relating to those those definitions. By the way, it is exartly the same same principle Korsakow, an editing software to make non-linear stories works (one scene is looking for another one which will continue the story, relying on 'who am I' and 'what am I looking for' definitions) - but that's the topic of anoher article here...
Another project in which Harris was a co-creator (again, with Sep Kamvar) is We Feel Fine - an exploration of human emotion. In it, a system searches for the sentence 'I feel' or 'I am feeling' in blogs and represents them graphically, according to different parameters.
Read an interesting interview with Jonathan Harris here.
The person who introduced me to the works of Jonathan Harris is Caspar Sonnen, the director of doclab.org, which is part of the International Documentary Festival in Amsterdam (IDFA). As of today, doclab is the first and only event which is solely dedicated to new mdia documentary projects.
Kaspar loves data visualization projects. So am I. But i admit that as a traditional documentary filmmaker, it took me some time to consider those projects documentaries. Luckily, kaspar is free from traditional convenstions and in his position in doclab has enough power and influence to set new rules regarding the definition, scope, borders and practices of documentary filmmaking.
true, it is not easy to recognize the 'story' as we know it in those projects. They don't have a beginning and end and as a viewer (actually, it's better to call it 'participant') you simply fall into an ongoing story when you enter the site, at a certain arbitrary stage from the point of view of the storyteller. Some of them never end at all. As those projects rely on raw data, some of them don't have protagonists to identify with or a plot - again, in the traditional meaning. In short, the models of storytelling, as formulated by Aristotle in Poetics, are destroyed one by one. Five doclab projects are now nominated for the Webby award - quite impressive, but how many documentary filmmakers heard about the Weby Award or aspire to be nominated to it? We still have a long way ahead, conceptually.
Finally, you're invited to watch an inerview about data visualization with Manuel Lima, an interaction desginer, information architect and design reseracher.
it's not about filmmaking, but it is inspiring: