I put a little new project online, analyzing which ingredients were ordered together most often in custom muesli mixtures from mymuesli.com customers. Besides chocolate, nuts and oat, the project description also features a little excursion on matrix views of networks, and conditional probabilities.

Finally, at least some documentation for the “Global Trade Flows” project I completed last year for citibank. Unfortunately, I cannot publish the full interactive tool, but I hope the screenshots are informative enough.

I started a podcast series — together with Enrico Bertini — named DATA STORIES. Exciting!

For the third year in a row, I was responsible for a good deal of the graphics in the annual Global Risk Report published by the World Economic Forum. For the report, hundreds of experts take part in a survey on their perception of what they consider the most important global risk and their inter–dependency.

Three types of graphics are at the heart of the report:

The “crystal” network diagram sheds light on the “Centers of Gravity” (systemically most important risks) in each category (Economic, Environmetal, Societal, Geopolitical, and Technological Risks) and the risks strongest connected to these. The network was layed out in d3.js, using force-directed layout and a “magnetic” grid for regular spacing and to avoid overlaps. The centers of gravity and the 4 most important connectors were fixated manually in this process, to enforce the “crystal” structure. This rough layout was then imported into Illustrator and refined and tweaked by hand. There is also a version which shows a cloud of all risks in the background, but I think this one obscures the conceptual/diagrammatic nature of the original, so personally, I prefer the cleaner version.

The same interconnectivity information can be explored in an “orbit” visualization that plays a bit on the gravity theme established in the survey. Clicking a risk will put it into the center and show how strongly the other risks are connected by how close or far away they are located — a very simple, but quite effective and clean approach to network visualization, by getting rid of the lines altogether and just working with size and distance to express connectivity.

Last, but not least, we have a simple cartesian plot arranging the risks by impact and likelihood. As we gathered some information on the respondents’ region of residence and their stakeholder group, you can explore how, for instance, Asian experts’ perception on economic issues differs from the rest of the respondents.

All interactive visualizations were implemented using jQuery, underscore.js, and raphael.js. For the network visualization, I used the force directed layout from d3.js.

Make sure to consume the full interactive report or in pdf form.

The fine folks over at Interactive Things interviewed a few great designers and hackers for their substratum interview series. I also answered some questions — read the interview here.

Demo and code for a smart little label placement method developed in the Max Planck research networks project.

For the younger ones among you: 5 years ago, infosthetics was pretty in purple (with that awkward off-center vertical axis). 5 years ago, there was no flowingdata.com. But yes, we did have electricity, thanks for asking.

What was I doing 5 years ago? Well, I was just starting my master’s at the interface design program at FH Potsdam. The fabulous Prof. Boris Müller took me under his wings, and he made me visualize the heck out of this brand new thing called web 2.0! Tags! Feeds! Blogging! Wikipedia! A new world order! Ah, exciting times. Although clearly a child of its times, I still think my thesis turned out quite nice, and laid the foundation for many, many good things to come. The weeks of my thesis presentation were probably the most intense of my life, with me and Sina moving together, me presenting my thesis, Sina’s father dying and our son Juri being born in the course of only a few very short weeks. A whole new chapter had started, and now I was living on the country side with a baby, commuting to Potsdam for my part-time research position, and at the same time working freelance datavis jobs.

Well, many things have happened in the meantime, and I am very grateful for where I am at right now. Here’s to the next 5 years of well-forming whatever data I come across!

I realized the project together with the talented Jan Willem Tulp, from who you surely will hear much more over the next few months.

A little technical note: The interactive graphics were all done in HTML5, with the help of the following great Javascript frameworks: raphael.js, underscore.js, d3.js. Except for the network visualization, all graphics run fully interactive down to IE6. #techachievement #lookma #noflash

You should definitely also check out the other videos in the eyeo collection, and watch out for upcoming releases, there were literally dozens of great, great talks.

The multi-touch installation Max Planck Research Networks reveals how Max Planck Institutes collaborate with each other, and with their international partners.

The piece is exhibited at the Max Planck Science Gallery, a highly interactive exhibition space presenting new forms of science communication in Berlin. You should come and visit it.

This is the first larger Java project I ever did, and it would not — at all — have been possible without Christopher Warnow from onformative, so thanks for for the great cooperation! (Big thanks go to the rest of onformative, too — to Cedric for the idea for the label placement algorithm and Julia for the great support with the documentation.)

Big thanks also to my colleagues at Elsevier and Max Planck Gesellschaft, who made the collaboration a real pleasure. I was also quite impressed with the overall project direction of Archimedes, who did a fabulous job of conceiving and realizing the exhibition space.