Truth and Beauty from Moritz Stefaner

Moritz Stefaner presented one of the most interesting and humorous views on the process of information visualizations, with nuggets of wisdom sprinkled in between variations on the design. 

From his introduction, “I would call myself a designer and the material I work with is code”, to his tips on working with data “Respect the data, pick out the interesting things, but let the data speak”, his talk was wonderfully satisfying.

His overview covered 4 projects:

1) Notabilia project - notabilia.net - a project visualizing deletion discussions on Wikipedia. 

His process was exploratory from the beginning, starting with an idea triggered by a discussion with other researchers and a large data set, the analysis of the data and then a variety of visual experimentation.  Key to finalizing the project, however, was the coincidence of the anniversary of 10yrs of wikipedia that led to making the final design decision - picking only 100 discussions from the original data set of thousands and using a single tree structure.

2) Revisit project - http://moritz.stefaner.eu/projects/revisit/ - the permanent tweet installation running throughout the conference.

It was interesting to hear that apart from the initial goal - Live dynamic visualization of tweets, the rest of Moritz’s pursuit was aesthetic, as he looked at ways to both feature the tweet content and create a more engaging form. 

3) Map Your Moves - really interesting process starting with maps, and extending through the analysis of complex reasons for moving to retain a general relative geographic location, but focus on scale comparison.

4) Better Life index project

The goal of the project was to allow users to see countries ranked by priorities such as income, jobs or life satisfaction. 

The process itself started with receiving a spec with the data, analyzing the structure of the data, using wireframes and then experimenting with graphing the actual data set in a variety of styles.  

In the end, an additional consideration was the relationship to the branding, and how well the image would stand up to represent the brand.

Most interestingly, the analysis and graphing didn’t stop with the launch of the project.  Instead, they continued to gather data from the way people used the site, what they engaged with, and which configurations were shared, and created a second, meta-visualization capturing these behaviors. 

Recap:

It is a capital mistake to theorize before one has data - Arthur Conan Doyle

“Tell a thousand stories, but not all at once.”