A couple of days ago, Vision Pro headsets started landing on people’s doorsteps. VR has been around for a while; this instance (which I have yet to experience) brings to the field Apple’s penchant for good UX, effectively creating a new way of interacting with computers. Expect other manufacturers to copy Apple’s approach, as usual.
Looking at reviews on YouTube, I’m reminded that interaction design never stands still. Surely, the “rectangles floating in virtual space” experience of the first-gen AVP is merely a bridge from familiar window-based interfaces to a more spatial computing-native experience. How do you design digital experiences unconstrained by the bounds of rectangular screens? Can you effectively do it with tools designed to present information in 2-D rectangles, such as Figma?
On another tack, consider generative AI. LLMs aren’t a more powerful version of an existing thing — they’re a wholly new substrate and material for digital experiences. What do you need to know to design with and for these and future disruptive technologies? I posit that the tools and methods that brought us here won’t get us where we want to go.
In my previous post, I speculated about five possible future roles for designers. The upshot: design work will look very different in the future. It’s inevitable, given the pace of technological change. How can you prepare for the future, whatever it holds? The answer is that you need to think differently about your work. Here, I’ll share three ideas I’ve adopted to remain relevant as a designer.
Change isn’t something to be feared but embraced. Don’t become trapped in the past by resisting new technologies and paradigms. Instead, experiment with new technologies to learn their limitations and capabilities — not second-hand, in the abstract, but in practice. This requires continuous learning and professional development — not just in the craft of design, but its purpose and orientation. What’s the ultimate goal of your work? What is it in service to?
Blacksmithing was a common occupation when horses powered much transportation. Horses needed shoes, so societies needed blacksmiths. This went on for centuries, with generation after generation of craftspeople doing the same things with the same kinds of tools to achieve the same ends.
The internal combustion engine mostly wiped out blacksmiths. While you can still find a few, smithing is no longer a mainstream occupation. Given how long it had been around, the change was astonishingly fast — a few decades. Some blacksmiths doubled down on smithing, while others became auto mechanics. Guess which ones had more of a future?
Intriguingly, blacksmiths had aptitudes that made them well-suited for working with cars. But first they needed to realize that their business wasn’t shoeing horses but enabling transportation. Designers must do the same. If you’re clear on what your work is in service to, you won’t be afraid of technologies that change the game.
Change is inevitable. Experiment with new methods, tools, and materials to see how they can allow you to work better, more effectively, or perhaps not at all. Don’t let romantic notions about craft or ideological convictions keep you trapped in an irrelevant occupation.
Ground Work in the Timeless
Embracing change is essential. But some things stay the same, no matter how much technology advances. Ground your work on things likely to remain relevant regardless of what changes.
Consider human nature. We are bipedal mammals with particular needs, capabilities, and constraints. We need food, rest, and companionship. We want to help others. We value accomplishment. We can also be greedy, needy, and complicated. Human psychology will likely remain in play regardless of how we interact with each other and the world.
The fundamentals of human societies are also remarkably stable. Although there is variation in forms and methods, most societies have been structured hierarchically, with elites (whether elected or not) governing everyone else. Division of labor (i.e., specialization) and collaborative value creation have also existed for a long time.
The old nurture, educate, and mentor the young, and the young care for the old. People trade goods and services. For convenience, they use tokens to represent value in exchanges. When disputes arise, they appeal to agreed-upon authorities for arbitration. In some cases, conflict escalates to violence.
Whatever else may change, these features of humans and their societies will likely remain constant. The critical question is: how can you use new technologies and methods to improve timeless things? Conversely, how can you create more effective experiences with new technologies, given the unchanging aspects of the human experience? What ethical considerations arise from the use of new technologies?
You can’t address any of these questions without understanding that which is unchanging. And the best way to do that is by reading old books. It’s no accident that people still read Marcus Aurelius, Lao Tse, and Shakespeare: their works express truths about human nature that have proven relevant to people in many different contexts. (Read up on the Lindy effect: things that have endured are likely to endure.)
The news won’t do it. Contemporary entertainment won’t do it. Social media certainly won’t do it. All are important and should play a role in your epistemological toolkit. But many are biased toward actions that benefit others (e.g., continued engagement, selling you crap, partisan politics) rather than what you want, which is ongoing relevance.
Adopt Models as the Object of Design
Let’s recap the previous two ideas. First, technology changes too fast for you to ground your work on current interaction paradigms. Second, human nature remains remarkably stable, regardless of how we interact with each other and the world. When considered together, you may deduce that it’s possible to develop timeless design skills by focusing on something other than UI. But what?
The answer is models. Hugh Dubberly has the best definition I’ve seen:
Models are ideas about the world — how it might be organized and how it might work. Models describe relationships: parts that make up wholes; structures that bind them; and how parts behave in relation to one another.
The idea that the Earth orbits the Sun is a model. You don’t have direct evidence for this, but that’s okay. The model has predictive power and is widely acknowledged as “mostly right.” Note that this wasn’t always the case. For a long time, most people believed the Sun orbited the Earth. Better instruments led to a different understanding. Models can change. (Not without effort.)
More importantly, models can be designed. The UI you’re designing reflects a particular understanding of what the app is supposed to be and how it’s supposed to work. Whether you’ve consciously defined a system model or not, the elements of the UI will lead users to develop a mental model of what the app does and how.
Your goal is to ensure they develop a good mental model — one that lets them do what they need using the system with relatively little effort. You can’t wing it when designing complex systems: you must define a model before drawing screens. You can (and likely must) change the model once you start drawing screens. But the model should lead.
Now, models are abstractions: most often, they’re represented as “boxes-and-arrows”-type diagrams. Abstractions can guide screen-level design, but can’t be used to validate directions. People read too much into abstractions. The only way to test your design directions is to render the model’s implications at the level of screens.
Here, “screens” is a metaphor. As I said above, you can’t fully predict future interaction paradigms. And eventually, AI may produce much of the UI-level design work anyway. In either case, you must understand how the system model maps to user interfaces.
Models are the central object of interaction design. (This includes, of course, interaction models.) Whatever design looks like in the future, designers will be more likely to define what systems are and how they function than what they look/feel like.
But you can’t fully define models without understanding how they affect UX at the interaction level. So you need a healthy relationship with abstraction: one that guides UI-level design executed by others (perhaps by AIs) and is informed by the tangible results of those high-level design decisions.
Toward Pragmatic Abstraction
You can think of this as a call for pragmatic abstraction: designing system structures independently of user interaction paradigms but practically grounded in improving user experiences (that happen, by definition, at the point of interaction.)
Intriguingly, this idea isn’t new or future-facing. Information architecture has focused on pragmatic abstraction all along; a sitemap is a type of conceptual model. Alas, IA’s reliance on abstraction is one reason it failed to reach mainstream adoption. Many people dislike abstraction.
Which is to say, shifting from screen-level design to model design won’t be easy. But as technologies evolve, affecting both end-user UI and the designer’s toolkit, designers must dive deeper to ensure continued relevance. That means making models the object of design to continue creating value by satisfying human needs — many of which will remain constant despite ongoing, pervasive change.