Life beyond the folder system

In 2021, the Verge published an article analysing the horror that the current generation of students couldn’t understand the way their professors thought about computers (Internet Archive version).

The core disconnect was around the folder system. It’s something so intuitive to generations of computer users that they can struggle to think of any other way of using a computer. They can even struggle to explain the folder system as a concept, because it feels so natural to them.

Yet for younger generations, the idea of a “file” and a “folder” is completely alien.

To understand why, we need to go back to where the folder structure came from in the first place.

Computer folder structures only emerged in the 1960s

Far from being “just how computers work”, the first operating system to offer a folder-like organisation system, Multics, only emerged in the 1960s.

The folder metaphor was developed in the decades following. It became a core concept of the graphical user interfaces that made home computing an accessible reality to everyday people.

This was successful because it was a metaphor for the way people worked in the 20th century. People working at real desktops used real folders and real files to organise their work.

It was vital to get this concept right. This 1984 YouTube video featuring Apple graphic designer Susan Kare shows her explaining the desktop metaphor, and how computer files and folders work, to a group of middle-aged men:

“If you want to store a file in a folder, it’s analogous to life. You just put the piece of paper in the folder.”

Of course, the file doesn’t actually exist in a folder. The folder doesn’t exist either. All those folders and files on your system only exist as digital data — zeroes and ones in a storage space. But that is too abstract for most people to work with.

Files and folders taught people to use desktop computers

Files and folders were the perfect metaphor to introduce computing to a generation of office workers. Metaphors are necessary in digital interface design.

Because computers work in abstraction from reality, interfaces effectively need to translate the computer’s work back into things the user can recognise from real life. This is what an interface is: the point at which the human and computer can understand each other.

The desktop metaphor was a masterstroke from Apple and other pioneers of graphical user interfaces in the 1970s and 1980s.

Smartphones needed new metaphors to make their vast capabilities learnable

A few decades later, Apple’s iPhone changed mobile phones forever. The interface was radical but intuitive. It also made very little use of the folder metaphor anywhere. It wasn’t even possible to organise apps into folders until iOS 4, three years after the introduction of the iPhone.

Instead, the interface concepts were skeuomorphic representations. Skeuomorphic is a fancy word that simply means: to mimic a real-life tool. This was needed in order to teach users how to use the range wide of functions that were suddenly made available in their pocket.

The camera app had an animated camera shutter that flickered when you took a photo. The compass app looked like an old-fashioned instrument you might see in a maritime museum. The recorder app looked like the sort of microphone George Orwell could have used for a wartime BBC broadcast.

At first, this worked really well as a way of teaching people how to use smartphones. It helped people understand the vast capabilities that were now on offer. This wasn’t just a computer in your pocket. It was everything in your pocket.

Skeuomorphic metaphors can become out-of-date quickly

But a few years later, everyone understood how to use a smartphone. When the interface concepts continued to cling on to those metaphors from the past, users began to feel infantilised.

The book app presented books on a graphical pine bookshelf of the sort you’d see in an underfunded library. The podcast app was presented as a reel-to-reel tape player.

People began to ask: “Why? I know what a podcast is.”

Others might have thought: “What is a reel-to-reel tape player?”

Taken too far, skeuomorphism began to amuse and confuse more than be of use. As a result, skeuomorphism, and metaphors in computing generally, became unfashionable for a decade.

If you’ve ever been dragged into a debate about why the “save” icon looks like a floppy disk of the sort not widely used since the 1990s, you’ve felt this problem in action. The save icon was the ultimate safety blanket for a generation of computer users. For the generation that followed, it offered only confusion.

This shows us that a metaphor that initially works brilliantly can still outstay its welcome. Even if the current generation of users get it, the next generation of users might be confused by it.

Different metaphors work for different generations

Files and folders have lasted longer than many metaphors in computing. It was a masterstroke from the golden era of graphical user interface design. It worked so well for the generation that started using computers in the 1980s, that the descendants of that generation were taught about files and folders as if that’s obviously just how computers work.

It didn’t matter that millennials made less use of real-life files and folders than their parents did. They understood the concept of the computer folder, because the desktop computers they grew up using were centred around it.

But the generations that followed didn’t learn to get stuff done using desktop computers. They learned to get stuff done on their mobile devices, where folders were seldom used. Not only do young people make no use of real-life folders, they make no use of computer folders.

Upcoming generations will understand different metaphors

I recently attended a webinar about the digital media behaviours of generation alpha, who are currently in their teens and approaching adulthood. Research suggests that people in generation alpha find websites harder to use than boomers do.

One hypothesis is that so much of generation alpha’s digital media consumption is centred around short-form video. This means that navigating an old-fashioned text-heavy web-based interface is something they have to learn.

Millennials were originally thought of as digital natives. But their experience of growing up with digital technology was nothing like what younger generations are experiencing. Put into that context, the experience of the original digital natives looks incremental rather than transformational.

After all, boomers could use technology as well. They used telephones and teletext at home, and many will have used computers or fax machines at work even if they didn’t have a computer at home until later in life. They just did all this in a way that met the capabilities of the technology, and their cultural expectations, of the time.

While millennials might have thought they were pioneering the use of digital technology, in reality they just found the way that best suited their generation — just like previous generations did. New generations are now pioneering their own new approaches.

Technology advances have rendered folders less useful

The Verge article pinpoints 2017 as the time professors started struggling to describe computer folders to their students. Those students will have got their first mobile devices just when smartphones began to take prominence over desktop machines.

It is also worth considering that millennials can remember a time before good search engines. They had to rely on strong folder structures, otherwise they might never have found their precious files again.

Nowadays, search is so powerful that not only can you find the file you need, but you can even find the specific part of the file you need — in the blink of an eye. Think of how Google takes you straight to the most relevant snippet of text on a webpage, or how an email generated by OneDrive can take you straight to a specific comment on a specific document. This power has made folders much less important as an organisation system.

In short: A generation that mostly consumes information as an infinite scroll of short-form video or instantly-findable snippets of text is not ready for your folders.

This does not mean that we should begin to publish all our information as short-form video. But it does mean we need to get smarter in how we use metaphors to structure our information.

We need to move beyond old-fashioned paper-based metaphors

Files and folders have outstayed their welcome. We’ve spent generations talking about files, folders and pages — then wondering why our digital services haven’t improved on the paperwork they replaced. That’s because we’re still using paper-based metaphors to describe it all. But the possibilities of digital are so much more powerful than that.

Going back to that article from the Verge, it is only towards the end that there is a glimmer of realisation that maybe, just maybe, it is Principal Skinner who is wrong.

But in doing so, the professors resort to an unhelpful “laundry basket” metaphor. That would rely too heavily on powerful search tools, and presumes that people will always know what they are searching for rather than seeking to learn what they need to find. It is also worth noting that search systems still rely on good semantic structures and metadata to return the right results, rather than magically plucking things from a digital laundry basket.

We need to move towards more object-oriented metaphors

Instead, we need information structures that more closely match the real-world things they represent and the real-world relationships between those things. This is the essence of the object-oriented approach.

Forcing everything into files, folders and pages warps the information to fit an old-fashioned paper-based mindset. That is inappropriate for the needs of the present generation, the capabilities of our current technology, and the nature of the information itself.

This is why we need to move beyond top-down site structures where information exists only as pages that must have one true home within a monolithic one-size-fits-all tree of folders. Instead, we need to think about the structure of the information for the things it actually represents — and the relationships between those things (which is not always parent–child).

It may seem radical to those who were brought up on the folder system. But the potential of something better has always been there.

After all, Tim Berners-Lee didn’t invent the world wide tree. He invented the world-wide web.