The Disappearing User
The term ‘user’ no longer reflects the complicated relationship between a person and a computer
The concept of a ‘user’ is fundamental for the design of complex digital systems. It can prominently be found in terms like ‘User Experience’, ‘User Interface’ and ‘User Centred’. Most designers will assert that understanding the user is essential for creating good design solutions.
However, I don’t think the ‘user’ still exists. I would argue that the established model of a self-determined human being operating a digital device no longer adequately reflects the relationship we have with digital technology.
I consciously say no longer, because the user did exist. For about twenty years — from the mid-80s to the mid-00s — human beings actually controlled and directly operated computers. Before that, computers were programmed, not used. After that, the relationship between computers and people radically changed. It turned from transactional to symbiotic. ‘Users’ no longer just use computers — computers also use ‘users’.
Our current model of interaction design is based on the dichotomy of user and computer. It describes the relationship between a person and technology. In this established model, the distribution of power and control is very clear: ‘users’ use and ‘computers’ compute. The human being is in charge and tells the computer what to do. The computer complies — as long as the commands are expressed in the right way.
However, this model is no longer accurate. It is still based on technical configurations from 40 years ago. If we look at the first Apple Macintosh from 1984, we had a fairly simple setup:
The computer (CPU and storage) was connected to the outside world in three ways. Input could be received via mouse and keyboard. Output was created via a monitor, a speaker or a printer. Storing files was only possible on physical floppy disks or external hard drives. While the first Macintosh already had a modem port, it was not conceived as a network computer.
Conceptually, the first Macintosh was very easy to understand. It was a contained system that very clearly conveyed its current state and the currently available interactions. Its entire functionality was based on the hard- and software that was included in the physical machine. The only way to expand its purpose was to manually insert a floppy disk with additional software.
In this particular setup, it makes sense to speak of a ‘user’. The computer itself was fairly passive. It waited for you to do something. It reacted and if there were things unclear, asked politely what exactly you had in mind. The state of the machine only changed when the user did something.
In a way, the first Macintosh was very obsequious. Like most simple machinery it only did something if you explicitly told it to do so. It was indeed operated by a user.
Fast forward to the newest iPhone:
This setup is much more complex. We can obviously still interact directly with the iPhone — we can manipulate elements on screen, take photos, or speak into the microphone. The results of our interactions are then displayed on the screen or piped through the speaker. But the power of modern mobile phones does not derive from touchscreens or microphones. It comes from the imperceptible side of the input / output dichotomy.
Modern phones have a secret life. They are packed with sensors and they are permanently connected to the internet. They are able to conduct numerous processes in the background without any interference of a ‘user’. Our phones are constantly exchanging data with online services. Sensor data and personal data is send to services — commands and external data are received from the cloud. The state of the iPhone is no longer determined by its owner but by a complex system of algorithms and data flows that are mostly invisible to normal ‘users’.
The fact that certain processes are imperceptible and run in the background is in itself neither new nor unusual. Abstraction, encapsulation and autonomous behaviour are at the very heart of modern computing and modern user interface design. Even the first Macintosh was highly abstracted. The illusion of the graphical user interface was only possible because a lot of technical processes were hidden from the user. But they were always directly invoked by the user. Opening a folder or moving a window around is simple for a person but fairly complex for a computer. (At least is was on the first Macintosh.) In this model, the user controls the state of the computer.
The state of a modern smartphone is not determined by a person — it is constantly negotiated by several autonomous processes in the background. The iPhone is permanently doing things on its own. Sometimes with our explicit consent — but quite often without.
It is important to note that the iPhone has no agency in itself. An iPhone has no intrinsic agency — it has not a mind of its own. But it definitely has a delegated agency. The algorithms that control the state were developed by human beings. And the creators and developers of the algorithms have indeed an agency. Depending on a set of parameters like sensor data and user data, they decide about the state of the iPhone.
Again — the simple fact that we are dealing with delegated agency is not necessarily a problem — it is an essential part of modern digital infrastructure. But it demonstrates that we are no longer dealing with contained systems and that even simple dialogues between humans and computers have become quite complicated.
Most users of a Macintosh did not understand the technology. The great thing about the Mac was that they did’t need to. They could write, edit, compose, draw without understanding the complexity of the underlying hard- and software. But the technology they did not understand was clearly defined. It was fairly easy to understand what they did not understand. Early Macintosh users were dealing with a known unknown.
Today, as we spend more time with computers than ever, we don’t know what we don’t know. We are dealing with an unknown unknown.
My problem with the term ‘user’ is not that it is inaccurate. My problem is that the term ‘user’ is suggestive. It indicates that a person is still fully in control of all the things that happen on their computer or their mobile phone. And I don’t think that is the case.
Our relationship with digital technology is constantly changing. We have been programmers, operators and users. Now, we are participants, contributors, consumers, clients. We are both actors and audience. But we are no longer users.
It is about time to come up with a new term that reflects the symbiotic relationship human beings have with computers. Old names like ‘Cybernetics’ come to mind. Or the more recent ‘Social Machine’. But both concepts emphasise the entire system and not the role of individual human beings with the system.
Even if we stick with ‘user’, I think it is necessary to re-evaluate the term and realise that its meaning has changed. The moment we interact with a computer, we are deeply embedded in a socio-technological setting that is way beyond our direct control.
User Experience Design and User Interface Design should reflect this. As designers, we have to acknowledge that the autonomous user is an illusion. We have to think more about systems, relationships and interdependencies.
We not only have to think about how human beings use technology. We also have to consider how the user is being used.
