'Information Carries Weight': How an Algorithm Helped Designers With the 9/11 Memorial

The final layout of the names around the South Pool, displayed in panels. color coding shows the affiliations between victims. [Image: Jer Thorp]

The subject line was unassuming: "Potential Freelance Job". The project itself, though, seemed nearly insurmountable--place the name of each victim of the 9/11 attacks on the Twin Towers into an arrangement that preserved both where people were and who they were with when they died, as well as a wide-ranging network of other relationships and attachments, within a set of panels surrounding the pools at the eventual New York City memorial. Digital artist Jer Thorp willingly took on the project, despite its complexity.

He went for a multi-level approach, of an algorithm-based rough layout with an interactive polishing step that let humans tweak the arrangement as necessary. The algorithm, designed using the Processing programming language, had to account for the names' spatial placement (affected by positioning around the two pools, length of the names, and typography), corporate affiliations (companies and departments), and personal adjacency requests that were filed by the families of the victims. The largest relationship clusters from these adjacency requests included more than 70 individual names--no small feat.

The algorithm was able to produce layouts that satisfied 98% of the constraints and adjacencies. Thorp's custom software tool took care of the rest, allowing designers and architects to manually adjust the computer-generated layout. Color coding displayed relationships, and several levels of magnification let designers view both fine-grained details and the overall effect. In addition, the software could export SVG files to allow for even smaller adjustments using visual editing programs like Adobe Illustrator. Thorp’s writeup on his website includes a video showing the tool in use.

Thorp, who intends to detail the workings of the algorithm at a later date, believes the project exemplifies computer-assisted design: “we should let computers do what computers do best, and let humans do what humans do best”. But in addition to that, he says that the design process was a strong reminder that all data is linked to real situations; while it may be easy to manipulate a data set, designers and researchers must remember the true source and meaning of the information they work with.

Check out Thorp's blog to read his full story on how it came together.

[blprnt via Flowing Data]

Other stories you might like...

Get your GeekTech on: Twitter - Facebook - RSS | Tip us off

recommended for you

Meet the GeekTechers

Read more »

Subscribe to the Best of TechHive Newsletter

Comments