I want to tell you a story about a map that never came to be. When I first heard it, in an interview with Katia Naouri in Desired Landscapes,1 its metaphorical power nearly knocked me over. The story itself appeared to me as a map of sorts, the destinations of which were almost innumerable. Very little information of it existed online. Digging deeper felt like an expedition in a way that so little research feels these days. I found myself traversing obscure foreign blogs, academic papers, and niche tourism pages. I biked over to an academic library, where I eagerly passed a French map anthology through an overhead archival scanner. I puzzled over translations, plot holes, and unexplained images. And I'm excited to share it with you.
The story goes something like this:
At the end of the 18th century, Spain's official geographer, Tomás Lopez, was asked by the King to create an accurate map of the kingdom. In an attempt to delegate the herculean labour required, Tomás drew a series of circles, picked the town in the center of each circle, and asked the local priest to answer a questionnaire and draw up a map of their province. The goal was to amalgamate the responses into a single map. But none of these priests were trained in cartography, and many of them would have had limited access to maps at all.2 Nonetheless, 500 of them tried. In one map, the entire region is represented simply by a series of letters (“A” for church, “B” for hermitage, “C” for house, “D” for tree, and so on). Another represents the surrounding villages as if they are orbiting planets. In some, the handwriting forms the topographies. In others, descriptive columns of text take center stage, as if the language itself is a landmark. Each priest implicitly reveals how they see the world around them, and the relative importance of its constituent parts: nature or people, religion or trade, architecture or landscape, precision or vibes.
Tomás tried for years to reconcile these mosaic shards with each other:
He died, exhausted, in 1802, after trying in vain to coordinate hundreds of maps that were inconsistent with each other. When, half a century later, the first Spanish geographical dictionary was published, it did not contain a single map. (B H Vayssiére, “Cartes et Figures de la Terre”)
There are many things to take from this story — about beginner's mind, the diversity of human experience, and the interoperability of language. But what stood out to me most was two opposing lessons about shared protocols and modularity. Tomás' experiment failed. It failed because each amateur cartographer injected their own methodology and process, resulting in incompatible maps. But in another sense, Tomás succeeded. Sure, maybe this collection of artifacts would be useless for military strategy or commerce, but on the other hand... LOOK AT THESE MAPS, THESE MAPS RULE. Imagining a world in which Tomás successfully imposed a protocol and stripped these maps of their individuality feels... tragic? Dystopian?
I'm obsessed with this story because it gets at a dynamic embedded within everything designed that we rarely think about. Once you notice it, it is present in almost every conversation, at every aperture and zoom level: modularity is inversely correlated to expressiveness.
If I were to design a personal map of my neighborhood, it would include the potholes I swerve by on my bike rides, the neighbor’s sweet precocious 4 year old that is always on the front stoop and wants to tell me about her day, routes that have small patches of grass to the right of the sidewalk and end near a public trash can (suitable for walking my right-side-only peeing dog), the schedule and trajectory of shade during the summer, homes with potted flowers hanging off their railings, restaurants that closed ages ago, the playgrounds where we are most likely to run into parent friend crushes, and the street with the best view of the skyline at night.
It probably would not look like this:
Let me give you another example. Close your eyes and picture the design of a news article on the web. You're probably imagining something like looks vaguely like this:
All of these stories, while sharing an interest in gastropods, have different subjects, tones, audiences, and takeaways. And yet, they look basically the same.
When you see a slightly generic news story design, you're seeing the solution to the problem: how do I make this work for any headline, any image, any tone, any audience? How do I make this as adaptable as possible? The question of “what does this story want?” becomes subsumed into the larger goal of malleability and efficiency. A template, not a story. A map, not 500 maps.
In an alternate universe, stories about snails could be much more visually diverse:
The downside, of course, is that approaching each story from scratch would require a lot more time and attention.
I don't want this to be mistaken for a story about web design. If you squint, you will find the same set of tradeoffs in all manner of creative work and systems thinking: think cookie-cutter suburban development, standardized testing, internet platforms, fast fashion, AI therapists, and multi-purpose kitchen tools. All of these opt for the flexibility of the generic over the expressiveness of the specific.
I am someone that preaches expressiveness to a fault, but the truth is that I make decisions to scale all the time. I don’t necessarily see this as a compromise of values. There is beauty in trying to express something specific; there is beauty too in finding compromises to create something epic and collective.
My only concern is whether we are considering the question at all. We have a whole host of platitudes for people in romantic relationships about give and take: “bend, but don’t break”; “compromise, but don't compromise yourself”; “meet halfway without losing your way.” But as participants in a designed world, it’s easy to forget that there is color lost when a system tries to be comprehensive rather than specific in the same way that a dish from a good local restaurant feels like a map of much more specific context than a TV dinner.
So when you see a map, ask yourself: how many maps could this have been?
☣️ Algorithmic detritus
Side effects of a world driven by software logic
Google's AI answers said to put glue on pizza, so of course Katie Notopoulos obliged: “No one with three brain cells would actually do this…but we can assume that AI also gives answers that are less obviously wrong but still wrong.”
Looking for oxy and escorts? Visit Eventbrite: “Eventbrite didn’t just publish these user-generated event listings; its algorithms appeared to actively recommend them to people through simple search queries or in ‘related events’… a search for ‘opioid’ in the United States showed Eventbrite’s recommendation algorithm suggesting a conference for opioid treatment practitioners between two listings for ordering oxycodone.”
🏃 Escape attempts
Acts of algorithmic resistance
&udm=14 is a tool for searching Google without generative AI answers
SmallStack is a newsletter that curates Substacks with fewer than 1,000 subscribers
The Tiny Awards celebrate “people making stuff on the internet for the fun of it and the love of it and the hell of it”
genderswap.fm is gender-swapped song covers
on scrolling is an infinite scrolling essay about infinite feeds
🌸 Screenshot garden
Souvenirs from my internet travels, presented without context
What would your personal map include? Where do you see one map where there should be 500? Let me know by replying, commenting, dming, or emailing me.
As one priest wrote: “I have not studied either the art of geography or cosmography. There are surely many faults in what you have asked of me; moreover, being a priest here for only three weeks, I am quite incapable.”
Thanks for this. I sincerely hope we’re headed to where we have the 500 maps as our default mode, and the ends to do with scale and modularity are met by machines. Effectively, the vain effort of Tomás, automated as appropriate.
Might be fun to get one’s hands on all 500 maps and see what Claude can do.
What a great story - thanks for sharing! Your commentary on what is lost with "scalable design" is something I think about a lot. There was a great TED talk by Jacek Utko about redesigning newspapers that is a fine example of this: https://www.youtube.com/watch?v=zHuH8P_Vqc0