Over the last week or so I was honored to participate in events at three institutions, which in some ways couldn’t have been more different: the Museum of Modern Art in New York City, metalab at Harvard, and the Bolinas Museum.
MoMA
Full Disclosure: The Edge of Information Design, MoMA’s first show devoted entirely to information design, opened at the Museum of Modern Art in New York City last week. Cabspotting, our project with Scott Snibbe and Amy Balkin, was one of the first data visualization projects that MoMA acquired, and I’m delighted to see it return to the galleries.
It’s hard for me to overstate the impact that this project had, and continues to have, on me and Stamen. For one thing, it provided an opportunity to not just put data on a map, but for the data to be the map. This simple gesture opened up a rhetorical space that I’m still exploring.
When our partner Eddie Elliot mapped thirty days of GPS data from Yellow Cab, he didn’t “correct” the places where it looked like the taxis were driving through buildings downtown. Instead he mapped the traces themselves, and in doing so opened up the idea that a fuzzy map wasn’t a source of error, but perhaps of a greater truth: that the fuzziness of taxi GPS could be a proxy for the heights of the buildings near them, since the satellites’ signals were bouncing off of the buildings before connecting with the taxis. So a fuzzy taxi map means high buildings where the blurs are. And, years later, while encountering Dietmar Offenhueber’s Authographic Design, he gave me the great gift of the idea that the height of skyscrapers downtown is a proxy for the nearness of the bedrock beneath. So taxi traces become proxies for the geology beneath the ground.
The project owes a debt to Laura Kurgan’s Information Drift, which placed a GPS sensor on the roof of the Storefront for Art & Architecture in New York and mapped the difference between the real latitude and longitude of the Storefront, and the coordinates a sensor was receiving from the scrambled-at-the-time satellite signals, which I was lucky enough to see before I ever knew Laura…
…so to be able to meet up with Laura (recent winner of the National Design Award) all these years later with both our work in the same show at MoMA was a real treat. Sometimes you get to party with your heroes, which I’m lucky to call Scott Snibbe, Paola Antonelli and Dario Calmese. Great fun!
Metalab
I’m lucky enough to have been invited to be a Senior Fellow at MetaLab, a global knowledge-design lab engaged in critical and creative practice across the disciplinary grid. Part of being a Fellow is giving a talk from time to time, and this time I presented on Archives, LLMs, data visualization, and where they connect. MetaLab is a terrific community of artists, architects, thinkers and makers, founded by Jeffrey Schnapp, and I knew I was in the right place when afterwards I found myself between a professional magician and an LLM researcher talking about alchemy and transubstantiation and sleight of hand.
| 
|
| 
|
What I like best about giving talks is that they force me to figure out what I’m thinking. What I learned from this talk is that many of the things I’ve been saying about data visualization…
- Don’t just use dataviz to eliminate uncertainty, but rather become better students of uncertainty and develop wisdom
- Don’t just use dataviz to measure accuracy, but also use it to measure drift
- Don’t just design for control, but also design for uncertainty
…are also things I want to say about AI! We spend all this time fretting over whether the results of our AI prompts are “accurate,” when (I think) it’s much more interesting to use them as a design medium.
Another thing I wanted to use the talk for was to put out a call about the next class I’m teaching at Harvard, Neural Cartographies: Mapping Brains and Machines, in the spring of 2027:
Neural networks, whether biological, artificial, or otherwise, are systems whose behavior emerges from large numbers of relatively simple components acting together. They are difficult to see, challenging to describe, and especially tricky to represent across scales. This workshop asks how design and visualization can make the structure and activity of neural systems perceptible without pretending to make them simple. We’ll look at LLMs, brains, and other neural nets and work to understand how these information landscapes respond to queries about space, meaning, power and landscape.
I’m starting to build out the class now, and I’m looking for partners in the class who can engage with us on these ideas and, crucially, provide some data for us to work with. If you know someone who works with brains or LLMs or other neural networks, and you’d like to see what engaging with a classroom full of smart talented Harvard GSD students can do with your data, please let me know!
Bolinas Museum
In what might seem like a departure from the usual Stamen material, but which I hope to stitch usefully into these discussions, I’m starting to talk publicly about a collaboration between myself and my friend and painter Nellie King Solomon. We’ve been working together for about a year. I’ve been making ink out of iPhones and guns, and Nellie’s been making paintings of them. This Saturday, one of Nellie’s paintings was shown for the first time in public, at the opening of the yearly auction at the Bolinas Museum. You can buy one!
| 
|
| 
|
I’m just blown away by the list of people we’re showing alongside: Mariah Nielson, Hughen/Starkerather, Barry McGee, Dave Eggers, these are all longtime heroes (and sometimes friends) of mine, and it’s an honor to be in the same room with them.
I’ll have more to say about the transmutation of iPhones and guns into inks soon. I love so many things about what Nellie’s doing with these alchemical materials, but what I think I love most is the obsessive documentation of the materials alongside their representation in the form of traced iPhones. Having worked with a number of museum curators, I’ve sometimes rolled my eyes when they talk about how difficult it is to track the provenance of objects in a systematic way, but having tried it myself, I’m learning how difficult it can be:
| 
|
More about all of this soon!