Can people live here?
Information architecture, read as a city.
Your version of the city
You don’t use the whole city. No matter how big it is, you move through the part that makes sense to you. Your home, your workplace, the gym, the two or three shops you always go to, your friends’ houses. That is your city. The rest exists, but to you it is little more than rumour.
Everyone has their own version. Some are broader, others narrower, each with its own conveniences and friction. And the more fluently someone can navigate their version, the more they feel they belong there. Belonging is, in large part, knowing how to move without having to think about it.
The same is true of any information space. No one experiences the whole thing. Each user carves out a path: the screens they visit, the things they save, the shortcuts they have learned by heart. A large app does not have one user. It has thousands of users, each living in a different slice of the same app.
That changes what the first job of information architecture actually is. The temptation is to organise the whole thing, to put everything in its proper place for some complete, imaginary user who can see the entire structure. That user does not exist. The real job is to organise for many partial, real versions at once, knowing that each person sees only a piece, and that we cannot assume those pieces will fit together into one perfect whole.
Jorge Arango makes this connection deliberately: thinking about information as a place, something you inhabit rather than a document you read. It is a useful lens because it changes the question. Instead of “is this structure correct?” you ask “can people live here?” A correct structure is easy to defend in a diagram. Whether a place is habitable can only be proven through use.
How space tells you where you are
If everyone inhabits their own version, the next question is practical: how do they find their way around? You arrive in a neighbourhood you have never seen before and, within minutes, begin to get your bearings. There is nothing magical about it. The city is telling you where you are.
Kevin Lynch gave names to the elements that make this possible. Paths, the routes along which you move. Landmarks, the points you use as references (the church, the tower, the strange-looking building). Districts, areas with a distinct character that you can feel yourself entering and leaving. And edges, the boundaries that separate one thing from another (the river, the wide avenue, the wall). A legible city is one where these elements are clear. You build a mental map almost without effort, and that map corresponds to the territory.
Information architecture does the same job with different materials. Navigation is a path. A persistent logo, a recognisable home screen, a consistent icon are landmarks. Sections with their own identity are districts. A shift in context between one area and another is an edge. When these things are clear, users know where they are, how to get back, and what to expect from the next step. When they are not, people get lost in exactly the same way they do in a city without reference points: going in circles, retracing their steps, giving up.
It is worth remembering that Richard Saul Wurman, who coined the term “information architecture”, was trained as an architect. The analogy is there at the origin of the discipline. Organising information has always been, at least in part, about designing a space that someone can inhabit and move through.
So far, legibility seems like an unqualified good. The clearer, the better. But there is still a question missing: legible for what, and in whose interest?
Spaces shape you back
So far, the story has moved in one direction: you build your version of the space, orient yourself, move through it. But the relationship goes both ways, and that is where it becomes interesting. You shape the space, and the space shapes you in return.
A neighbourhood without pavements is not neutral. It makes you drive. A neighbourhood where shops are far away from housing decides that you depend on a car to buy a loaf of bread. Nobody forces you, and yet the structure has already made a choice on your behalf before you arrive. What is close, what is far away, what is easy and what takes effort: these things govern behaviour more powerfully than any instruction sign.
The same thing happens in information spaces, only it is less visible. What sits on the first screen and what is buried five taps deep. What comes pre-selected and what you have to go looking for yourself. The order in which things appear. None of this is the user deciding. It is the structure deciding for the user, and users will almost always follow the path of least resistance. Every default is a decision made on your behalf by someone who is not you.
And this is where the city analogy helps us see something that it cannot explain on its own. Why is the space arranged this way rather than another? Because it carries two kinds of value at the same time, and they do not always point in the same direction. For you, the space has use-value: you want to find, do, save, solve and leave. For whoever owns the space, it has exchange-value: your attention can be sold, your behaviour can be collected, your time there can be converted into revenue. Much of interaction design lives in the tension between those two forms of value.
A feed is the clearest example. It presents itself as a service: information organised for you. But its architecture is optimised to keep you scrolling, because time spent there is what gets sold. Both things coexist on the same screen. When use-value and exchange-value align, great, nobody notices the seam. The problem begins when they diverge, and that is the point where information architecture stops being a technical question of organisation and becomes a question of who it serves.
That is why the asymmetry matters. Yes, you shape the space: your data, your clicks, your behaviour train the system, feed the rankings, adjust what you see. The relationship goes both ways, that is true. But the two sides do not have equal power. You shape the space one click at a time, without a plan and without seeing the whole. Whoever owns it shapes the space deliberately, with the entire map in front of them and a clear objective. Calling this “co-creation” sounds nice, but it is a lie. It is a power relationship dressed up as convenience.
Cities are contested. Apps aren’t.
This is where the analogy begins to break down, and where that breakdown becomes most useful.
We have arrived at an uncomfortable place: privately owned space is a one-sided relationship of power. But there is a fair objection. Cities are not fair either, they have owners too, they also serve those with power. So what is the difference? If both are unequal, why should the city be a useful model for thinking about information? The difference is not inequality. It is what can be done against it.
Cities are not democratic either. People with more money live better, move around more easily, receive better services. Urban planning serves particular interests, rarely those of the people with the least power. None of this is new. But the city has something that privately owned information spaces do not: it is contested.
Henri Lefebvre called this the right to the city. The idea is that a city does not belong only to those who hold legal title to its property. It also belongs to the people who inhabit it, use it and make it function every day. And because it belongs to many people, it becomes a permanent site of contestation. Residents fight for a pedestrian crossing. A neighbourhood resists being turned into a car park. An abandoned space is occupied and claimed. Over time, the city accumulates the results of these struggles, shaped by many hands, with no single one able to dictate the whole. Part of what makes a city habitable is precisely that nobody has complete control over it.
Privately owned information spaces are not quite as absolute, but they tend in the opposite direction. Think of it as a spectrum. At one extreme is the social media platform: one owner defines the structure and users have almost no recourse. You do not get to contest the order of the feed, negotiate the default, or call a public hearing about the hierarchy of the menu. Your options are to accept it or leave, and leaving is not having a voice, it is giving up. In the middle sits most apps, particularly those that depend on paying users. There, people have some voice: feedback channels, the pressure created by cancellations, customer support, and the ways people actually use the product, which inform (or should inform) design decisions. It is a weaker form of contestation than what exists in the street, without assemblies or any right to occupation, but it is real. At the other extreme sits the city, where people have the right to stay and fight.
This is where the city becomes useful as an allegory for designers. It does not only teach us how to organise a space well. It shows us what may be missing from that space. The city becomes the benchmark: a good place to inhabit is usually one that many people can shape and contest. No app will ever be a city. What matters is knowing where it sits on the spectrum, and how much voice you can return to the user. An app can be fast, beautiful and clear and still have more in common with a gated community than with a city.
Planning a city is different from planning a building
There is one last difference, and it is the one that matters most for anyone working at scale.
Designing a building is a closed problem. You know the brief, you design the spaces, you decide where everything goes. In the end, the building is more or less what was on the plans. A screen works like that. You decide what goes where, and the result is what you designed.
A city does not. Nobody designs a city building by building. Planners define something else: where there can be shops and businesses and where there can only be housing, how high people are allowed to build, where roads can run. Thousands of people then build within those rules, and the city that emerges is never exactly the city that existed on paper. The planner does not control the outcome. The planner controls the rules that produce the outcome.
Jane Jacobs spent her life criticising planners who forgot this, those who looked at the city from above and tried to design the whole thing as though it were one enormous building. It does not work, because what makes a city live is the people who use it, on the pavement, every day, in ways no top-down plan can fully predict. The mistake is not planning. It is confusing planning a city with designing a building.
This is one of the hardest shifts for a designer to make, and I see it up close. Designing a screen is designing a building. Designing a platform where other teams build is planning a city. You stop positioning every object yourself and start defining the rules: what other people are allowed to build, which pieces they can use, and within which boundaries. A design system is zoning. A governance framework is land-use law. You do not decide what every screen will look like. You decide who has the right to decide, and what the space allows.
And this is where all the threads come together. Defining the rules of a space is the strongest act of power within it, stronger than any individual screen, because those rules shape everything that comes afterwards. If organising information means building a place for people to inhabit, then the question that remains is not whether it is well organised. It is who it serves, and how much voice belongs to the people who are going to live there.