In praise of friction

The value of learning curves and context in UI/UX design

Josh's guitar, a 1958 Gretsch Hollowbody in a shade of green called "Avocado".
Photo courtesy of Josh Fischer

"Easy" and "intuitive" are words you hear an awful lot if you make websites, apps, or any kind of interface. Over the last 20 to 30 years, they’ve become watchwords of user interface (UI) design; our metric for a design’s success is often based on how self-evident it seems to unfamiliar users. Can anyone, anywhere, pick it up and start using it without needing outside help? Does it "just work" for most people?

User experience (UX) designers often describe their approach as a process of reducing "friction." We spend a lot of time trying to figure out what our users’ goals are, and then do our very best to eliminate any "friction points" that slow them down. Along the way, a certain minimalism (and sometimes even asceticism) has overtaken UI design, as we strip away anything that feels even potentially complicated or unpredictable.

"Perfection is finally attained not when there is nothing more to add, but when there is nothing more to take away," said [Antoine de Saint-Exupery](https://en.wikipedia.org/wiki/Antoine_de_Saint-Exup%C3%A9ry), and we cling to this thread like we’re dangling over an abyss. And maybe we are.

But I’ve been realizing lately that if we look outside the world of UI/UX design, with a broader view of what an interface is, the "easier = better" formula sometimes breaks down. Interfaces can be found everywhere; they’re hardly limited to screens. I can imagine UI problems to solve in many other fields:

  • industrial design: buttons, labelling, textures, packaging
  • automotive design: ergonomics, control layout
  • architecture: accessibility, floor plans, furniture layout
  • tool-making: handles, weight, safety
  • civil engineering: traffic flows, congestion
  • urban design: crosswalks, bike lanes, signage, zoning, parks
  • fashion: fit, fabric, closures

And so on. Arguably, any object or environment that you expect humans to use for a specific purpose (and/or have a specific experience) can be seen as an interface. UI, by any other name, is still UI. By this definition, there are interfaces I use every day that don’t meet the standard of being "easy" that, nonetheless, I still find valuable. There are, in fact, many cases where a certain interaction feels all the more meaningful and rewarding to me not in spite of its difficulty but because of it.

I love playing vinyl records, even though I can just ask Alexa to stream the same album with much less effort and (arguably) better fidelity. But the process of pulling a record from the shelf, putting it on the turntable, and setting the needle down feels gratifying to me in ways I have a hard time explaining. There’s something about bringing home a physical album and then deliberately playing the whole thing that forces me to "invest" in the music in a way that, I believe, makes me appreciate it just a little more. There is undeniably more "friction" in my experience with vinyl but it somehow enhances the experience.

I like writing and drawing with fountain pens instead of newer ballpoints or felt-tips or digital styluses, even though the latter are undeniably easier to use. When I got my first fountain pen, in fact, I spent weeks scratching holes in my paper before I learned how to hold it properly. I like the way fountain pens creates a less "perfect" line that can be modulated with pressure and speed. I like the happy accidents and blots that sometimes appear. It’s a slower, less efficient way of making lines, but, again, the effort feels satisfying to me rather than frustrating.

These cases point to certain assumptions underlying the way I’ve usually approached my design work that seem worth questioning. Is easier really better all the time? When is a learning curve actually worth it?

Friction and inspiration 

Musical instruments are an interesting example of "hard-but-worthwhile UI." Let’s put on UX goggles a moment and imagine designing an ideal, "frictionless" instrument. First, consider what a musician is basically trying to achieve when she plays: to make a series of notes, or sounds at precisely controlled pitches. Taking that as my brief, I would set about searching for the easiest way for our musician to produce pitched sounds. Pitch is really just a tone at a certain wavelength, so I might eventually land on something that looks a lot like an oscilloscope. We can use a single dial to adjust pitch — just turn it to change the wavelength! Super simple! We can add a display to tell her exactly what note/wavelength she’s playing, since precision is a requirement here, and positive feedback always reduces friction. Our instrument could be mass-produced as a compact electronic device, or even sold as an app for tablets. Nailed it, right? But no: instead of my simple, easy oscilloscope, many musicians continue to insist on creating their music with anachronistic nightmares like the piano, the guitar, the cello, the drum, or — God help us — the oboe.(Note: I realize my "ideal design" is actually similar to the virtual instruments many electronic musicians and DJs use these days: folks who often perform entire concerts on laptops or mobile devices. A lot of amazing new work — and amazing new sounds — are being produced this way, but it’s interesting to me that electronic instruments have evolved to _coexist_ with traditional ones, rather than simply disrupting or replacing them. Most musicians I know work with both, e.g. playing their guitars into laptop-based samplers, and feel that tradition and technology complement one another in their work.)

Unconscionably complicated and awkward, most of these instrument designs literally injure the people who use them. They have no labelling or displays to tell you what notes you’re playing. Instead you must spend a lifetime learning to "hear" the correct tones and to produce them with your hands or mouth or both. Instruments are usually heavy, oddly shaped, ergonomically terrible, and fiddly. Most of them require constant maintenance and tuning, are impossibly sensitive to humidity and climate, and are constructed, by hand, from deliberately arcane, hard-to-source materials that must be so precisely fitted that master craftspeople spend decades training apprentices to pass on their knowledge. Many instruments have been produced exactly the same way for hundreds of years, with almost no significant innovation. Why, then, do musicians subject themselves to such difficult, friction-filled interactions? In this case, of course, the difficulty is a feature, not a bug. A piano is hardly "intuitive," but the process of learning to play it — and developing the range of skills needed to play expressively — is something that musicians find deeply rewarding and ultimately a source of inspiration. Over time, the piano’s odd controls and layout have become not a hindrance but an important _context_ for creating and understanding music. And, in turn, musicians have found all kinds of interesting new ways to create sounds with them. Whoever invented the piano couldn’t possibly have foreseen Thelonious Monk or Arnold Schoenberg or Alicia Keys, but it was the piano — with all its quirks and pain points — that provided the sound they each transformed into a voice all their own.

History and familiarity

I was visiting my friend Josh recently when he motioned me over to a corner: "I got myself a Christmas present."

And he lifted a guitar off its stand and held it out to me: a gorgeous 1958 Gretsch Hollowbody, in a subtle, pale green called "avocado." He told me about how he’d seen it at a store, fell hard for it, and then agonized for weeks about buying it. One of the more expensive guitars on the market in 1958, the Hollowbody has remained valuable and desirable ever since.

After he finally decided to bring it home, he agonized another few weeks about how much he’d spent on something so hard to (at least rationally) justify. He’s not a professional musician, and he’s never had a band. He’s just an enthusiast who wanted to work harder on his music, and he found inspiration in that beautiful guitar.

"It has a soul," he said. "I lose track of time and reality. I played for 6 hours the other night without realizing it."

It’s in nearly perfect shape, but you can see some of its 60-year history on its surface; the coat is a little less shiny in some places from where it’s been held. It makes you wonder about all the hands that have played it over the years. To Josh (and, I imagine, many musicians), that history is part of what he loves about it. Guitars have evolved since 1958 but Josh’s Gretsch is just as usable today as it was new (or, perhaps, even better). There’s a certain timelessness about a carefully-made, beautiful instrument that becomes part of the experience of using it. Josh can already imagine passing it on to his kids and grandkids. And perhaps that’s something he thinks about every time he picks it up; it’s a voice that others can remember him by, and one that can be passed on so that it becomes something shared between generations.

From a certain, big-picture perspective, the most successful interfaces are the ones that have stuck with us the way that pianos and guitars have, the ones people have a hard time letting go of even when easier or simpler or newer solutions become available.

Looking at it this way, I’m realizing how terribly important context is to interface design. UX designers need to understand our users’ goals, of course, but we also need to have a deep, intimate understanding of their history, traditions, and point of view. Context creates the language with which your users describe their goals, the metaphors by which they understand them, and therefore what feels familiar and "easy" to them. As John Raskin, a computer scientist at Apple, said in a now-famous paper, "it is clear that a user interface feature is ‘intuitive’ insofar as it resembles or is identical to something the user has already learned. In short, ‘intuitive’ in this context is an almost exact synonym of ‘familiar.’"

Originally, graphical operating systems like Apple’s OS and Microsoft’s Windows were built around a set of metaphors that they imagined their users — presumably people who’d worked in an office before — would already be familiar with. A process stored to disk was a "file." A group of files was a "folder." Folders were accessed on your "desktop." There was a "trash can" for deleting them. These GUI mechanics were roundly touted as "intuitive" but if you’d never seen a file cabinet before, all those icons would be meaningless to you.

To someone new to Western music, a piano might at first look like an impossibly opaque user interface, but for musicians familiar with its context — and the "language" of scales and octaves it uses — it’s not only intuitive but is perhaps the basis of their approach to their craft. Its difficulty is ultimately rewarding, rather than frustrating, in part because of its deep history and the worldwide community that has developed around it. When you sit down at an old grand piano and strike a key, you can’t help but feel a connection to all of that.

As a designer, it’s easy to get swept up in the cult of the new. The desire to reinvent is in our nature; our industry moves fast and technology creates new possibilities almost every day. It’s also easy to get stuck in a mindset where ease and speed and innovation should always overrule familiarity and tradition and history. But, in the end, these are the things our craft is built on. There is no intuitive without familiar and there is no new without history. What looks like friction, from one point of view, might in fact be capability that only reveals itself with time and understanding.

I keep a fountain pen on my desk to remind myself of that.