Information architecture and metaphor

Sarah R. Barrett
Known Item
Published in
6 min readApr 11, 2023


I’ve written previously about architectural skeuomorphism, that is, the attempt to make digital experiences structurally resemble physical locations (tl;dr, it doesn’t work, don’t do it.) I’ve also written, more recently, about metaphors to explain why we do information architecture. Metaphors are inescapable in information architecture, not just to explain things, but because each generation of technology inherits the language of what came before. A great example of this in action is how the folder metaphor is breaking down across the generations, as detailed in “File Not Found” by Monica Chin.

Students in technical fields are having trouble using file-based software:

[Catherine Garland] asked each student where they’d saved their project. Could they be on the desktop? Perhaps in the shared drive? But over and over, she was met with confusion. “What are you talking about?” multiple students inquired. Not only did they not know where their files were saved — they didn’t understand the question.
Gradually, Garland came to the same realization that many of her fellow educators have reached in the past four years: the concept of file folders and directories, essential to previous generations’ understanding of computers, is gibberish to many modern students.

When I teach, I use examples from grocery stores to explain wayfinding principles. It works really well. Grocery stores organize hundreds of thousands of products, but you’re never shocked to be able to go into an unfamiliar one and find a specific brand of whole wheat bread. (Amazingly, you can even usually go into a brand new grocery store and be sure that they don’t have the thing you’re looking for, which is the gold standard of findability to me. We almost never get there in digital experiences.)

Aisles in a typical American supermarket
A typical American supermarket has predictable paths, identifies its locations, progressively discloses information about products, and provides landmarks.

The grocery store metaphor works wonderfully. People almost immediately make the jump to pointing out additional things grocery stores do to improve wayfinding.

You know where it doesn’t work? Europe. Urban European grocery stores are totally different, and they don’t do any of the same things, so they’re no good for teaching wayfinding.

Urban supermarkets in Europe tend not to identify aisles, often have shorter sight lines, and fewer landmarks.

What makes metaphors work

Metaphors work for teaching or improving usability when a familiar, concrete source domain’s characteristics have significant overlap with a new, abstract target domain.

American grocery stores don’t help teach wayfinding principles in Europe because they aren’t familiar to a European audience. Urban European grocery stores don’t help teach wayfinding principles in Europe because they don’t have enough overlap with the principles to illustrate it well. To go a different direction, color theory might be familiar to a group of people, but it won’t help teach wayfinding, because it’s not concrete enough.

For a metaphor to work, you’ve got to have all three: Familiarity, concreteness, and overlap.

Architectural metaphors

When you use metaphors to explain the information architecture of digital experiences, users make assumptions about the affordances of a system based on the metaphors it uses. This is a folder? Clearly, things can go in it. Clearly, I can’t listen to it.

For these metaphors to increase usability, though, they need a high degree of overlap between the source domain and the target domain. Inevitably, things in digital experiences cannot do everything their metaphorical sources can. I can’t take notes about the contents of my downloads folder directly on it, like I can with a manila folder. My downloads folder also does many things that the physical object cannot. It can be copied! I can make a shortcut to it! All these affordances are left out of the metaphor.

File folders

In this case, professors and students are coming to a central computing metaphor from different places. For professors, the file folder metaphor is familiar and concrete. The metaphor effortlessly suggests affordances and limitations that the system stands by. Professors have used computers with directory structures for so long that the metaphor itself has become familiar and concrete, and I’m sure you could teach these professors any number of abstract concepts by comparing them to a computer directory.

The problem is, for these students, the metaphor is neither familiar nor concrete, so it cannot be used to explain how things work on a computer. And in the article, you see professors soliciting new metaphors:

[One professor] put out a call for useful analogies on Twitter and was met with various suggestions: physical tree branches and leaves, kitchen utensils sorted into drawers, books and shelves in a library, “Take their phones away and get ’em on Windows 98.”

The problem is, kitchens or books are not the metaphors encoded into the designs of these programs. These are metaphors to explain a metaphor, all furiously trying to justify fundamentally arbitrary constraints.

The truth is, digital files do not have to take up one and only one spot on a shelf or in a drawer. The metaphor helps to communicate some of the limitations of a system, but it’s not intrinsic to digital experiences. Like all good information architecture, this is a shared fiction. That “shared” part is breaking down, weakening this approach.

Lastly, it’s worth recognizing that this shared fiction is breaking down for a good reason. It embodies many limitations that our experiences no longer adhere to. A system that surfaces what you need when you need it, rather than keeping it from you until you supply an exact file path, is delightful and generous. The problem, according to professors in this article, is that programming languages don’t work this way. It’s crucial to add, though, that this is a latency in our technology and its metaphors, not an inherent characteristic of it.

Beyond files

Our insistence on an architectural metaphor of outdated technology limits us. As Ted Nelson said about 1999 technology [PDF], hewing too close to the metaphor of paper is like “tearing the wings of a 747 and driving it as a bus on the highway.”

Given how long smart people have been talking about this, it’s not happening rapidly, but it is happening. Experiences are moving in this direction, and you can see the metaphors falling away from younger users, despite educators’ best efforts:

Plavchan now also spends a lot of time teaching his students about directory structure in his courses, along with other basics, like file extensions and terminal navigation. Guarín-Zapata begins his semesters with a similar tutorial. “I start with a little talk about a mental model of a computer, what a computer is,” he says. “We have memory; we have a hard drive; we have an interface; we have a file structure.”

Modeling the new, “vertical” method of filing papers. From the Classified Illustrated Catalog of the Library Department of Library Bureau (Boston: Library Bureau, 1900), pg. 113

The vertical file, the basic technology of the modern day filing cabinet, was cutting-edge information technology of 1898. Together with the typewriter, carbon paper, and other associated technologies, it replaced spindles, pigeonholes, and pressbooks. In all systems, papers were organized chronologically, as they had been since our earliest libraries.

Pigeons in pigeonholes (left), documents in pigeonholes (right).

For those of us who are considerably older than Google, can you imagine if you spent significant time in school learning about an 18th century technology based on the idea that most people knew what pigeon housing looked like?

Metaphors are a powerful tool in explaining and improving the usability of digital experiences, but they tend to be rooted in the technologies of the past. Understanding that won’t help undergrads find their files, but it might help us build a better generation of tools.