Five Design Essays Worth Returning To
Pieces about how the work actually gets done, between the brief and the finished thing.
This is a working list rather than a canonical one. The five pieces below are the ones I keep coming back to when a project has gotten stuck, when a decision needs recalibrating, or when I notice I've started designing on autopilot. Some of them are essays in the strict sense. One is a chapter, one is a list, and two are books I treat as essays because their central arguments live in a single chapter each.
1. "A City Is Not a Tree" by Christopher Alexander (1965)
Alexander's essay is short, technical, and slightly mathematical, which makes it easier to skip on a first pass. The argument worth slowing down for is in the distinction he draws between two kinds of organizational structure. A tree is a strict hierarchy where every element has one parent and elements never share children. A semi-lattice is messier, with overlapping membership and elements that belong to multiple sets at once.
Alexander's claim is that natural cities are semi-lattices and planned cities are trees, which is why planned cities feel sterile and natural cities feel alive. The semi-lattice contains the kind of cross-cutting connections that produce real life, and the tree's insistence on clean separation kills exactly the thing that makes a place worth being in.
I read this essay roughly once a year, usually when I'm tempted to clean up a design system, a content taxonomy, or an organizational structure that feels messy. The piece is the reminder that messiness is sometimes the load-bearing feature rather than the bug to be fixed.
2. "The Design of Everyday Things," Chapter 1, by Don Norman (2013 revised edition)
This is technically a book recommendation, but the first chapter does enough work on its own to count. Norman lays out his vocabulary for thinking about how objects communicate their use: affordances, signifiers, mappings, feedback, and the gap between designer intention and user understanding. Everything else in the book elaborates on what Chapter 1 establishes.
The version of the book worth reading is the 2013 revised edition rather than the 1988 original, which Norman rewrote after the iPhone changed the design conversation. The chapter is the one I return to whenever I catch myself drafting an explanation that exists because the design didn't communicate something it should have.
3. "Ten Principles for Good Design" by Dieter Rams (developed across the 1970s and 1980s)
Rams developed his ten principles over roughly two decades at Braun and Vitsoe, mostly as a working philosophy for his own design practice rather than as a public manifesto. The list reads as deceptively simple. Good design is innovative, useful, aesthetic, understandable, unobtrusive, honest, long-lasting, thorough down to the last detail, environmentally friendly, and as little design as possible.
Each principle is more demanding than it looks on a first reading. "Honest" alone disqualifies most contemporary product design. "As little design as possible" is the principle most violated by the people who quote it most loudly. The list works best as a periodic audit. Read the ten principles, hold a current project up to them, and notice which ones the work is failing.
4. "The Web's Grain" by Frank Chimero (2015)
Chimero gave this as a talk before publishing it as an essay, and the talk-roots show in the rhythm of the writing. His argument is that the web is a material with its own grain, and designers spend much of their time fighting that grain rather than working with it. The metaphor is precise. Wood has a grain. Paper has a grain. The web has flexibility, fluidity of layout, variable viewport, and a natural pull toward responsiveness. Designs that fight those properties tend to feel brittle. Designs that work with them tend to feel inevitable.
The piece is the one I return to when I find myself wrestling a layout into a fixed shape it doesn't want to take. The metaphor generalizes outside web design. Every medium has a grain, and most design problems get smaller when you stop fighting the grain of the material.
5. "How Buildings Learn" by Stewart Brand (1994)
Brand's book is about architecture, but its central idea is a model that applies to almost any complex designed system. His shearing-layers framework describes a building as six layers that change at different rates: Site (geological time), Structure (decades to centuries), Skin (about twenty years), Services (seven to fifteen years), Space Plan (three to thirty years depending on use), and Stuff (daily and weekly).
The point is that good buildings accommodate the different time-scales of their layers, while bad buildings lock them together in ways that prevent each from changing on its own clock. A building where the wiring is buried inside the structural concrete is a building that has fought against its own grain, to borrow Chimero's frame from the previous entry.
The model has been borrowed extensively in software design, where the distinction between fast-changing UI and slow-changing data layers maps onto Brand's framework directly. I return to the book when I'm designing something that has to last long enough to change, which describes most things worth designing.
The five pieces don't argue for the same thing. Alexander wants you to preserve messiness. Norman wants you to take responsibility for user confusion. Rams wants you to remove rather than add. Chimero wants you to work with the material. Brand wants you to design for change over time. The reason all five sit on the same shelf is that they describe different aspects of the same underlying practice, which is the work of making something that holds up when it leaves the studio.
Get methods in your inbox. Subscribe