The Kiosk We All Know
There was a time when waiting for a train meant learning the geography of a kiosk. José Fidel remembers it with more affection than he expected.
You could stand there for forty minutes without buying anything and still become strangely familiar with its entire universe. Newspapers overlapped one another behind glass, a few glossy magazines occupied the privileged territory at eye level, cigarettes formed their stern little architecture behind the seller, and somewhere between chewing gum, batteries, postcards, pens, chocolate, lottery tickets and objects nobody remembered needing until they saw them, a complete miniature economy managed to fit inside a space barely larger than a wardrobe. The remarkable thing was not how much the kiosk contained. It was how little space it needed in order to feel sufficient.
A supermarket made the kiosk look absurd by comparison. Hundreds of metres of aisles, dozens of varieties of water, shelves devoted to sauces, pasta shapes and household products in numbers no single person could reasonably compare, refrigeration systems humming for things that might never meet the particular customer entering that day. The kiosk could not compete with such abundance and never tried. It had no room for abundance. Every object had to justify its centimetres. Yet on a station platform, with eighteen minutes before departure and rain beginning to move sideways across the tracks, that tiny window could suddenly become more useful than an entire shopping centre.
That is the first Kiosk Principle: abundance and usefulness are not the same thing. A system can become more useful precisely because it cannot contain everything.
Why Does a Kiosk Sometimes Beat a Supermarket?
The supermarket answers a broad question: What might anyone possibly want? The kiosk answers a much narrower one: What are you likely to want here, now? The difference appears trivial until we notice how much of modern technology has moved from solving the first problem toward solving the second.
The early internet was intoxicated with access. More pages, more files, more songs, more images, more information. Search engines became maps to abundance. Streaming removed the shelf. Cloud storage removed the cupboard. The smartphone absorbed objects that once occupied separate drawers, rooms and institutions. For a while the direction of progress seemed obvious: make more available, remove more barriers, expand the catalogue until scarcity of information disappears.
Then the bottleneck moved.
Having ten films is a limitation. Having ten thousand films is a menu problem. Having one encyclopedia means there are facts you cannot reach. Having a search engine connected to billions of pages means the difficulty shifts from access to judgment. Which answer matters? Which article deserves ten minutes? Which message deserves a reply now? Which photograph should be kept? Which of the almost infinite possible things a connected person could do deserves this particular hour of a finite life?
The supermarket solves scarcity of stock by increasing stock. The kiosk solves scarcity of time by reducing choice. That may be why the kiosk survives conceptually even when many of its original functions disappear. Its strength was never inventory. Its strength was compression through relevance. A small space forces an intelligence onto the shelf. Someone has already decided that on a cold platform you are more likely to want coffee than a chandelier.
And, once upon a time, that someone was visible. She sat behind the glass. She knew which newspaper was under which pile, where the batteries were hidden, whether the new issue had arrived and which magazine a customer preferred not to ask for too loudly. You did not search the kiosk. You asked the kiosk keeper. The warehouse could remain obscure because one human being had learned its geography.
This was curation before we needed the word. The choice at the window had already been reduced by wholesalers, publishers, shelf space and the judgment of the person behind the counter. It was restrictive, certainly. It was also strangely honest. Nobody standing before a kiosk believed the little window contained the whole world.
What Does Limited Space Teach Us About Interfaces?
Smallness is not automatically wisdom. A kiosk can stock rubbish. A limited menu can omit the one thing you actually need. Reduction can become poverty just as easily as elegance. But physical constraint introduces a pressure abundance can avoid. When there is room for only twenty visible objects, those twenty objects must compete differently from twenty objects buried among twenty thousand.
This is where the kiosk begins to resemble a good interface.
A good interface does not expose the whole machine. It exposes the part of the machine that is useful at the point of contact. A television remote does not display the electronics of the television. A map application does not begin by showing the mathematics of satellite positioning. A banking app does not greet the customer with the architecture of settlement networks, authentication protocols and clearing systems. Civilization increasingly works by hiding enormous structures behind surfaces small enough for ordinary human attention.
The kiosk did this physically long before software did it digitally. Behind one small window sat publishers, factories, distributors, printing presses, tobacco companies, confectionery manufacturers, postal systems, lotteries, transport schedules and supply chains. None of that was visible to the person buying a packet of gum. The simplicity at the front depended on complexity behind it.
Children occasionally noticed a different mystery. Where did the kiosk lady go to the toilet? The question is funny because it breaks the interface. For a child, the woman behind the window could seem almost like part of the kiosk: face, hands, voice, newspaper, change. Then one vulgar little question restored everything the window had hidden. She had a body. She needed a break. She had arrived from somewhere and would eventually go home.
There is a curious reversal here. We once turned a person into an interface because the counter hid most of the person from us. Now we often do the opposite: an interface speaks fluently and we begin imagining a person somewhere behind it. The old kiosk keeper really did have a life beyond the window. An AI does not become human merely because the window has learned to talk like one.
This is one of the most persistent misunderstandings about technological progress. We imagine that convenience means simplification of the world, when it often means only simplification of the contact point. The world underneath may be becoming more complex than ever.
Complexity does not disappear. It moves behind the interface.
The kiosk is a tiny physical demonstration of that rule. The shelf is simple because the network behind the shelf is not.
What Happened When the Smartphone Ate the Kiosk?
The smartphone performed an extraordinary act of cultural cannibalism.
It absorbed the newspaper, the map, the timetable, the calculator, the portable music player, the address book, the diary, the camera, the postcard and much of the magazine rack. It also absorbed a large part of the erotic material that once occupied its own socially awkward territory behind the kiosk window. Information migrated into the screen because information travels easily between carriers.
Fidel misses one thing about the old arrangement. Not the cigarettes, not the cramped booth and not the poverty of choice. He misses being able to see the person who was choosing. The kiosk keeper stood in front of the mechanism of selection. The smartphone made the counter cleaner and the warehouse immeasurably larger, but the keeper did not simply disappear. Ranking systems, recommendation engines, platforms and algorithms inherited part of the job. The difference is that the new keeper is much easier to forget.
By that logic, the kiosk should have died.
Instead, it changed.
You cannot download a hot dog. You cannot stream a bottle of water. A photograph of coffee does not deliver caffeine, and a digital chocolate bar may be infinitely reproducible while containing exactly zero calories. When information left the kiosk, the kiosk moved closer to the body. Food, drink, temperature, immediate need, physical convenience: these became more important precisely because the informational layer had migrated elsewhere.
That gives us a useful structural rule:
When information becomes digital, the kiosk retreats toward appetite.
This is not failure. It is adaptation.
Technology often removes one layer of scarcity and reveals another underneath. We once needed the kiosk for news; increasingly we need it because the body is hungry at 10:47 p.m. We once bought maps there; now we may buy coffee before consulting a map on the phone. The function changes, but the structure remains remarkably stable: a tiny interface presents a small number of immediate possibilities at the exact point where a person is likely to need them.
The carrier changes.
The structural problem survives.
Why “More” Is Not Always the Same as “Better”
A hypermarket is magnificent when the task is provisioning a household for a week. It is absurd when the task is buying water before boarding a train. That distinction exposes a deeper error in the way technological progress is often imagined: we assume that the superior system is the system capable of doing more.
Capability matters, but only relative to circumstance.
A machine with ten thousand functions can be worse than a machine with three if the user needs one of those three immediately and cannot find it. An AI capable of generating one million plausible solutions to a problem may be impressive in a benchmark and intolerable in practice. The first hundred answers may feel like abundance. The millionth may feel like paralysis.
At that point, the valuable intelligence is no longer the intelligence that can produce another option. It is the intelligence that understands why only three options should be placed on the counter.
This may become one of the most important shifts in how we think about AI. The first phase celebrates generation: more text, more images, more plans, more simulations, more alternatives. The next phase may be about restraint: selecting what deserves attention, hiding what does not, preserving access to the warehouse without forcing the human user to walk through every aisle.
The future may not need a larger interface.
It may need a better kiosk.
Why the Luxury of the Future May Be Not Seeing Everything
There is something almost indecent about the modern amount of choice.
A person can spend an evening moving between films without watching one, scroll through restaurants without ordering, compare hotels until every hotel begins to seem disappointing, save articles without reading them and keep photographs that will never be opened again. The old problem was scarcity. The newer problem is that possibility itself can become friction.
This does not mean abundance is bad. Scarcity is rarely charming to the people actually suffering from it. Access to more medicines, books, jobs, foods, educational resources and sources of information is one of civilization’s genuine achievements. The important distinction is between having access to abundance and being forced to confront all of it every time a decision is required.
A library containing ten million books is wonderful. A table containing ten million books is unusable.
Civilization increasingly requires mechanisms that preserve the first while preventing the second.
The kiosk is therefore not an argument against abundance. It is an argument for layers. The warehouse may be enormous. The counter should remain human-sized. A good system allows the deeper catalogue to exist without demanding that the user process the whole catalogue at once.
This is where selection becomes an ethical problem as well as a technical one. Who decides what appears? What disappears? What deserves eye level? A physical kiosk makes this visible because the shelf is obviously finite and the keeper is sitting behind the glass. A digital system can hide the act of selection entirely. The better the filtering becomes, the easier it is to forget that filtering is happening.
Perhaps the old kiosk was more honest. It never pretended to contain the whole world. And perhaps that is what Fidel misses: not the smaller choice, but the visible responsibility for making it smaller.
That tension may define the next generation of interfaces.
Could the Future Be a Counter Rather Than a Warehouse?
The Kiosk Principle becomes more interesting when we move beyond information.
Suppose future technologies allow richer changes to perception, attention or experience. Not the crude science-fiction fantasy of simply downloading another person’s complete memories, but smaller and more plausible extensions: new sensory mappings, temporary perceptual modes, immersive environments, perhaps new ways of experiencing information that currently require years of training.
The usual instinct is to imagine an infinite catalogue. Every environment available. Every perspective selectable. Every possible sensory state stored somewhere for immediate access.
But that instantly recreates the supermarket problem at a more terrifying scale.
What do you choose when there are a million possible afternoons?
Perhaps the mature interface would not resemble an infinite library at all. Perhaps it would resemble a kiosk: a few possibilities placed in front of you because they fit your location, mood, intention, available time and current curiosity. Twenty minutes experiencing colour through an unfamiliar perceptual mapping. An hour encountering orbital mechanics spatially rather than symbolically. A temporary environment designed around music, architecture or scale. Not everything. Something.
The warehouse behind the system could be effectively infinite.
The human standing at the counter would still be finite.
That asymmetry may survive every technological revolution.
Infinite warehouse. Tiny counter. One customer. Limited attention.
The Kiosk Principle does not predict a specific device. It identifies a constraint that technology cannot simply wish away. However much becomes possible, experience still has to happen somewhere.
For now, somewhere is a person.
Watch the Video
How to Read the World
by Jose Fidel
Current Knowledge
Human decision-making is strongly influenced by how choices are structured, but the popular idea that “more choice is always worse” is too simple. Larger choice sets can improve outcomes when options differ meaningfully and the person has enough information to evaluate them. Problems arise when alternatives become too numerous, too similar, poorly organized or difficult to compare. Context, expertise and stakes all matter. Modern interface design responds to this through ranking, filtering, defaults, recommendations, progressive disclosure and menu structures that reveal complexity only when it becomes relevant.
A kiosk provides a useful physical analogue because its limitations are visible. Limited space forces inventory selection, and usefulness depends heavily on context: location, time, weather, customer flow and immediate need. A hypermarket offers breadth; a station kiosk offers proximity and speed. Neither is universally superior because they solve different forms of scarcity. The same distinction applies to digital systems. A massive underlying database may be desirable, while an interface exposing the entire database at once may be unusable.
Digitization also shifts value from access toward selection. When information becomes cheap to copy and distribute, attention, trust, curation and relevance become comparatively more important. This does not mean access has stopped mattering or that information itself has lost value. It means that solving one bottleneck creates pressure at another point in the system.
Artificial intelligence may intensify this shift. Systems capable of generating or retrieving vast numbers of plausible outputs will increasingly need mechanisms that decide what not to show. The technical achievement may therefore move from production toward meaningful reduction.
Open Questions
The most important unresolved question is who should perform the selection when systems become too large for any individual to inspect. A shopkeeper selecting products is visible. A recommendation algorithm is not. As interfaces improve, curation can become both more helpful and more powerful while becoming less noticeable. The user may experience only the result, not the structure that produced it.
A second problem is the status of the intermediary. The kiosk keeper was visibly human and therefore visibly limited: she could forget, misunderstand, recommend, refuse or simply be wrong. Digital curation can perform the same structural function while hiding both the selection process and the identity of the selector. Conversational AI adds a further complication because a non-human interface can use human language well enough to invite anthropomorphic interpretation.
A third problem is serendipity. A kiosk’s limitations occasionally produced accidental discoveries: you came for a newspaper and noticed a strange magazine, a paperback or a ridiculous pen. A system that predicts preference too accurately might reduce these encounters. Perfect relevance could paradoxically make the world narrower by showing us only what fits our existing profile.
There is also the question of how far the kiosk analogy can be extended beyond goods and information. Experiences, identities and states of perception may prove much harder to package than media. A hot dog is a product. A state of consciousness may not be. Any future technology built around experience would have to confront questions of consent, identity, reversibility and psychological consequence that ordinary retail metaphors cannot solve.
Finally, there is the deeper issue of whether human beings actually want optimal selection. Part of culture comes from wandering, delay, error and curiosity. A perfect kiosk may be efficient. It may also be boring. The most human interface may need to know not only what belongs on the shelf, but when to place something there that does not.
Transhumation Interpretation
The kiosk matters to Transhumation because it demonstrates that civilization does not always progress by exposing more complexity. Often it progresses by compressing complexity into a smaller point of contact.
A legal system becomes a form. A bank becomes an icon. A city becomes a blue route on a screen. A library becomes a search field. The structures are not gone. They have become easier to carry because the interface has become smaller.
The danger is that what becomes invisible can also become difficult to question. The old counter showed that selection was taking place and even showed you the person performing part of it. The opportunity is that a good interface can make an impossibly large system human-sized.
Perhaps that is the real lesson of the kiosk.
Not that we should want less.
That we can only encounter the enormous through something small enough to stand in front of.
FAQ
What is the Kiosk Principle?
The Kiosk Principle is the idea that a limited interface can become more useful by selecting from a much larger underlying system. A kiosk cannot contain everything, so its value depends on showing what is relevant in a particular place and moment.
Is the argument that less choice is always better?
No. More choice can be valuable. The argument is that access and presentation are different problems. A large catalogue may be useful while a small interface makes that catalogue usable.
Why compare a kiosk to digital technology?
Because both hide complexity. A few visible options can sit on top of very large networks of infrastructure, rules, supply chains or data systems.
What does the smartphone have to do with the kiosk?
The smartphone absorbed many informational functions once handled by kiosks and other physical places: maps, newspapers, timetables, photographs, messages and magazines. The physical kiosk then shifted toward things a screen cannot directly provide, such as food, drink and other bodily needs.
Could AI become a kind of kiosk?
Yes, in a structural sense. As AI systems generate or retrieve more possibilities than a person can inspect, one of their most important functions may become intelligent selection rather than simple production. The analogy should not be confused with personhood: the old kiosk keeper was a human being temporarily functioning as an interface; a conversational interface does not become a human being because it performs some similar tasks.
What is the deeper Transhumation question?
If future civilization gives us access to enormous amounts of information, experiences and possible worlds, the problem may no longer be what exists. It may become what deserves to be placed on the counter now.
The warehouse may be infinite. The counter should not be.
You do not need everything. You need enough.
When access becomes abundant, relevance becomes the scarce resource.
Continue the Transhumation Series
- Forgotten Religion | Ancient Gods, Symbols and the Hidden Structures of Human Civilization
- The Last Religion | Death, Immortality and the Future of Humanity
- The New Theurgy | AI, Information and the Future of Human Civilization
- End of Branding: You Were Never Meant to Become a Product
- Baal vs Ishtar: Why Ancient Gods Never Disappeared