A curated collection of books, long form articles, and sites worth your time.

Each of the hundred entries is two or three pages: a specific finding from cognitive or perceptual psychology, the study behind it, and a quick note on what it means for the thing you’re designing. No grand theory, just a large pile of well-sourced, individually useful facts.

Good as a reference to flip through rather than read straight, and a solid way to back up a design decision with something more concrete than “it felt right.”

Written in 2008, well before Patreon or Substack existed, but it predicted almost exactly the economy those platforms would end up running on. The math is simple - 1,000 fans spending $100 a year is a viable living - but the reframe, depth over reach, is the part that actually changed how people think about audience.

Worth reading now specifically because the platforms it predicted now exist, so you can check the theory against a decade and a half of real evidence instead of taking it on faith.

Kelly wrote this for his birthday - one bit of advice for each year he’d lived - and it reads like it: no hedging, no caveats, just the compressed version of what he wishes someone had told him earlier.

Not every line lands the same for every reader, which is sort of the point. It’s a list built for returning to at different points in life, not consuming once and filing away.

Horton and Quesenbery build the book around seven personas covering different disabilities and situational limits, then walk through how real design decisions land differently for each of them. It reframes accessibility as a design problem with real users, not a legal checklist run at the end of a project.

The most useful shift the book makes: accessible design keeps turning out to be better design for everyone (captions help in a loud room, clear contrast helps in bad lighting), not a tradeoff against it.

This is the one you reach for after the shorter, friendlier books have already made their case and you want the actual mechanics: how personas are supposed to work, what goal-directed design means in practice, and a long catalog of interaction patterns with the reasoning behind each one.

Not a cover-to-cover read. It’s a reference that rewards being opened to a specific chapter when you’re stuck on a specific problem - forms, navigation, mobile gestures, whatever’s in front of you that week.

Kelly’s list of eight “generatives” - immediacy, personalization, interpretation, authenticity, accessibility, embodiment, patronage, findability - was written for the music and publishing industries mid-collapse, but reads like a direct explanation of why people still pay for things the internet makes free to copy.

The core move is reframing scarcity: once a copy is worthless, value has to come from something that can’t be copied at all.

Kholmatova frames a design system less as a Figma library and more as a shared vocabulary a whole organization has to agree on and keep maintaining - the patterns matter less than the process that keeps them consistent as the team and product both grow.

Drawn from real case studies (Airbnb, Eurostar, and others), which keeps it grounded instead of theoretical. Useful even if you’ll never build a formal system yourself, since most of the failure modes she describes show up informally on any team past a certain size anyway.

Krug’s whole pitch fits in the title: if a user has to stop and think about how something works, it’s already broken. The book is short on purpose - he says so himself - and every page is built around getting you to actually go run a usability test instead of just reading about one.

I hand this to anyone new to product or design before anything denser. It’s the fastest way to build the instinct that most usability problems are obvious once you watch one real person get stuck.

Garrett’s five-plane model - strategy, scope, structure, skeleton, surface - is one of those diagrams that escaped the book it came from and now shows up in decks everywhere without attribution. Reading the original is worth it just to see the argument in full instead of the compressed slide version.

Short and a little dated in its web-era examples, but the underlying structure (each layer depends on the one below it, and skipping ahead always costs you later) is still exactly how good product work gets built.

Müller-Brockmann was one of the architects of Swiss (International Typographic) design, and this is his case for the grid as a discipline rather than a constraint - a structure that makes a layout more legible and more consistent, not less expressive, once you actually learn to work within one.

Bilingual (English/German) and printed the way you’d expect a book about grids to be printed - precise, restrained, no wasted space. It’s older than most of this list but the underlying logic is exactly what still governs every responsive column system in use today.

Eyal’s Hook Model - trigger, action, variable reward, investment - is a clean, mechanical description of how habit-forming products actually work, and it’s genuinely useful as a diagnostic tool once you see it. You’ll notice it everywhere afterward, including in your own phone.

Worth reading with a little skepticism. Eyal has walked back some of the original framing over the years given how the same loop gets used for attention-extraction as often as for something people are glad to have a habit around. Read it for the model, argue with the ethics yourself.

Covert’s definition of a “mess” is broad on purpose: a messy closet, a confusing nav menu, an overloaded spreadsheet all count, and the same basic moves (naming things clearly, grouping them sensibly, deciding what to leave out) apply to all of them.

Written in plain, short sentences with almost no jargon, which makes it the book I hand to someone who’s never heard the term “information architecture” but is drowning in one anyway.

Cagan’s central complaint is that most companies run “feature teams” - groups handed a roadmap of things to build - instead of real product teams empowered to solve a problem however works best. The book is basically a long argument for the second model, backed by decades of consulting across companies that do it well and companies that don’t.

It reads a little like a manifesto in places, but the distinction between output (features shipped) and outcome (problems actually solved) is one of those ideas that’s obvious once stated and hard to un-see afterward.

A quieter, more nuanced companion to The Design of Everyday Things. Norman’s argument here is that complexity isn’t the enemy - confusion is. A washing machine with twenty settings isn’t automatically worse than one with three; what matters is whether the complexity maps to something the task actually requires.

Worth reading specifically as a corrective to the “just make it simpler” reflex, which doesn’t always hold up once the underlying task is genuinely complicated.

Victor’s central claim is uncomfortable if you’ve spent your career on interactivity: most software would be better as well-designed graphic design - context-aware information display - than as something you have to operate at all. Interaction should be the last resort, not the default.

Long, dense, and worth every minute. It’s less a set of tips and more a full reframing of what “good software” is even trying to do, and it changes how you look at dashboards, forms, and settings screens afterward.

Kelly coins “protopia” as an alternative to both utopia (arrival at perfection) and dystopia (arrival at collapse) - a state of continuous, incremental improvement where new problems are expected to show up right alongside the solutions to the old ones.

It’s a more honest and more useful lens for tracking progress than either extreme, and a good corrective for anyone whose sense of the future swings between utopian hype and doom.

Most design books stay conceptual. This one is almost entirely tactical: use fewer font sizes than you think you need, don’t rely on gray text for hierarchy, add depth with layered shadows instead of borders, that kind of thing. Each tip is small enough to apply in the next five minutes.

Written by and for people who can already build a working interface but keep landing on something that looks unfinished. It’s the fastest route from “functional” to “looks like a real designer touched it” I’ve come across.

Borrowing Brian Eno’s term, Kelly argues that what looks like lone genius in hindsight was usually a dense, supportive scene - a group of peers trading ideas, competing, and building on each other’s work in the same place at the same time.

A useful corrective to the myth of the solitary genius, and a good argument for why proximity to the right people still matters even when everything can happen remotely.

Ryan Singer wrote up the actual process Basecamp uses to ship product

• six-week cycles, an “appetite” set before scoping instead of an estimate after, and a betting table instead of an ever-growing backlog - and Basecamp gave the whole thing away for free online instead of selling it.

The appetite-over-estimate idea alone is worth the read: deciding how much time a problem is worth before you scope the solution, rather than scoping first and hoping the estimate comes in under budget.

Knapp lays out the design sprint hour by hour, day by day: Monday mapping the problem, Tuesday sketching solutions, Wednesday deciding, Thursday building a fake, Friday testing it on real people. It reads like a recipe because it’s meant to be run exactly as written the first few times.

The specific five-day structure isn’t sacred - most teams that use this end up adapting it - but the underlying discipline (get to a testable prototype before debating opinions any further) is the part worth keeping.

Dieter Rams wrote these ten principles in the 1970s while running design at Braun, and nearly every “principles of good design” list written since has been paraphrasing them without saying so. Reading the original is short and worth doing just to see the source.

“Good design is as little design as possible” is the one that gets quoted, but the whole list holds up - honest, unobtrusive, understandable, long-lasting. Ten short lines that still function as a genuine test for whether something is actually well designed.

Norman’s core argument hasn’t aged a day: when people struggle with an object, the object failed, not the person. Doors that need a sign telling you which way to push are bad doors. The same logic scales up to software, dashboards, and every interface with a confusing default.

It’s the closest thing this list has to a first-principles text. Read it once for the vocabulary (affordances, signifiers, mapping, feedback), then keep it around as a reference for arguing with anyone who says “the user should just read the instructions.”

Originally a talk, transcribed with Victor’s usual density of ideas per paragraph. The argument: humans think with far more than words on a screen - gesture, spatial memory, physical manipulation - and most software has narrowed “thinking” down to typing and reading because that’s what’s convenient to build, not because it’s what’s humane.

More provocation than practical guide, and better for it. It’s the kind of piece that reframes what you think the ceiling for software even is, rather than telling you what to change tomorrow.

Robin Williams (not that one) built the whole book around four principles - contrast, repetition, alignment, proximity, memorized easily as C.R.A.P. - and then spends the rest of the pages showing the same flyer or resume redesigned until each one clicks.

It’s aimed at people with no design background at all, which is exactly why it works as a gut check even if you have one. If you can’t explain why a layout is or isn’t working using these four words, you probably don’t actually know why yet.

Structured like a field guide: 125 concepts, one per spread, each with a short definition, an illustrated example, and a note on where the idea originally came from (psychology, architecture, industrial design, wherever). Hick’s Law sits next to the Von Restorff effect sits next to progressive disclosure.

Not a book you read start to finish so much as one you keep nearby and pull off the shelf when you need the actual name for a principle you’ve been applying by feel for years.

Kelly’s answer to the specific anxiety of feeling like you missed the window - that the internet, or whatever the current wave is, already had its founding moment and you weren’t there for it.

His counter is that right now is always the earliest possible moment relative to what’s coming next, which is less a comforting platitude than an argument backed by how early every previous wave actually was when it looked already-obvious in hindsight.

A place to collect and connect things - images, links, text - into channels that build on each other over time, without an algorithm deciding what surfaces. The opposite of a feed, in a good way.

A Roman emperor’s private notebook, never meant to be published, which is exactly why it holds up. No system to sell, no audience to perform for. Just a man reminding himself, page after page, to do the work in front of him and let go of what he can’t control.

I come back to this one every couple of years. It reads differently each time depending on what I’m dealing with.

Lays out how few times you actually have left to do the things you assume you’ll always have time for. Visits with your parents, seasons with your closest friends, all of it looks different once you count them out on a calendar instead of feeling them out. Genuinely changed how I plan a year.

Short enough to read in one sitting, dense enough to reread every year. Musashi’s point isn’t really about swordsmanship - it’s that mastery in one discipline teaches you how to see every other one.

The single best argument I know for why biography can be more instructive than any strategy book. Genuinely long, genuinely worth the time - budget a season, not a week.

Long, but Chernow earns every page. The most useful thread is how Rockefeller’s public and private selves diverge - ruthless in business, almost monastic at home - without either side canceling out the other.

Greene’s case studies (Darwin, Temple Grandin, a handful of lesser- known masters) are more convincing than the framework wrapped around them. Best read as a set of biographies with a shared thesis, not a productivity manual.

Less a how-to than a how-it-works. Read as description rather than prescription, the historical case studies behind each “law” are the actual value - the laws themselves are just the index.

Frankl’s core claim - that meaning can be found even in suffering, and that finding it is what actually gets people through the worst of it - is the rare big idea that survives being stated plainly. Short, and heavier than its page count suggests.

Jason Kottke has been curating the interesting corners of the internet since before “curation” was a word people used. Still the best proof that a simple link-and-a-sentence format never needed to be reinvented.

Tyler Cowen and Alex Tabarrok post short, dense links most days - less a blog you read cover to cover, more one you dip into for a different angle on whatever’s in the news.

A slower, more thoughtful corner of the internet on how to think rather than what to think. The mental models series is the best entry point.

Ben Thompson’s daily analysis is the clearest writing on tech business strategy I know of. Worth the subscription just for the aggregation theory posts.

Colder than its reputation suggests, and more useful as a study of power than a guide to using it. Worth reading once for how bluntly it says the quiet part out loud.

A framework for the difference between following a process and understanding the thing well enough to write a new one. Changed how I think about the difference between good execution and real judgment.

A friend’s correspondence you get to eavesdrop on centuries later. Best read a letter at a time, not straight through.

On why some effort compounds and some doesn’t, and why that changes what’s actually worth optimizing for. Pairs well with “How to Do Great Work” from the same author.

A transcript of a talk, not a polished essay, and better for it. Hamming’s question - “what are the important problems in your field, and why aren’t you working on them?” - is the kind that’s uncomfortable in a useful way.

Emily Wilson’s translation is the one to read if you haven’t - clean, propulsive, no dust on it. A homecoming story that’s really about patience.

Long, but the rare essay where every section earns its length. The bit on noticing what you can’t stop thinking about has stuck with me longer than almost anything else on this list.