There's a restaurant in East London that serves four dishes. Not four categories. Four dishes. Total. You walk in, you sit down, and you eat what they've made that day. No substitutions, no specials board, no separate menu for allergies and preferences and the vegan friend who decided to come along at the last minute. Four dishes. Take it or leave it.
It's one of the best meals I've ever had. Not because of what was on the plate, but because of everything that wasn't. Every decision had been made for you. Every ingredient had earned its place. Nothing was there to fill space or tick a box or hedge a bet. Four dishes, and every single one of them was perfect.
That is editing. And editing, I've come to believe, is the highest expression of taste.
The cult of more
We live in an age that worships addition. More features, more options, more content, more choice. Software products launch with feature lists that read like instruction manuals. Restaurants have menus the length of novellas. Streaming services have catalogues so vast that you spend more time choosing what to watch than actually watching anything. The assumption baked into nearly everything we build and consume is that more is better. That abundance is a virtue. That giving people everything they might want is the same as giving them what they need.
It isn't. It is the opposite.
The paradox of choice is well documented at this point, but it goes deeper than decision fatigue. The real problem with more is that it's a signal of something else entirely: indecision. When you can't decide what matters, you include everything. When you don't have the conviction to choose, you let the user choose instead and call it "flexibility." When you aren't sure what your product is, you make it everything and hope the market figures it out.
That isn't generosity. It's cowardice dressed up as comprehensiveness.
The courage to cut
I had a list of over forty ideas at one point. Forty things I wanted to build, ranging from the practical to the absurd to the genuinely ambitious. A daily briefing tool. A restaurant recommendation engine. A taste-scoring system. A bookmarking app. A wearable tech aggregator. A culture dashboard. A Japanese language learning tool. A pub guide. A parenting directory. A personal finance dashboard. On and on and on.
I built fourteen of them. The other twenty-nine - and this is the important bit - are the reason the eleven are any good.
Every idea I chose not to build freed up attention, energy, and taste for the ones I did. Every feature I chose not to add to Modern Retro kept it focused on the one thing it does brilliantly - AI-generated images of brands reimagined in a 1970s retail aesthetic. Not brand comparisons. Not historical context. Not user-submitted brands. Not a social layer. Not a marketplace. Just the images. Just the concept. Just the thing that makes you stop scrolling and stare.
That restraint isn't easy. Every week, I think of something that could be added. A rating system. A voting mechanism. A way for users to request brands. And every week, I choose not to. Because the moment you start adding, you start diluting. You go from a thing with a clear identity to a thing that's trying to be everything. And things that try to be everything end up being nothing.
Trove was the same. The temptation to turn it into a full bookmarking app was enormous. People understand bookmarking. It's a familiar category. Adding folders, tags, sharing, collaboration - all of it would have made Trove "bigger." But bigger isn't the point. Trove is a taste engine. It looks at what you save and tells you something about who you are. That's the entire product. The moment I add collaboration features, it becomes Notion. The moment I add social sharing, it becomes Pinterest. The moment I add folders, it becomes every bookmarking tool that already exists. The editing - the things I refuse to build - is what makes Trove, Trove.
Less but better
Dieter Rams said it decades ago: "Less, but better." Three words that contain an entire design philosophy. Not just less. Less, but better. The reduction isn't the goal. The improvement that comes from reduction is the goal. You remove things so that the things that remain can be excellent. You cut so that what survives has room to breathe, to be noticed, to matter.
Think about the albums you love most. Really think about them. Are they the ones with twenty-two tracks, where six or seven are brilliant and the rest are filler? Or are they the ones with ten tracks, where every single one belongs? The best albums are edited. The best restaurants are edited. The best wardrobes are edited. The best portfolios are edited.
The music analogy is useful because it makes the principle visceral. Nobody listens to an album and thinks "I wish there were more mediocre tracks in the middle." But that's exactly what we do when we add features to products. We pad the tracklist because we're afraid of the silence. We're afraid that fewer features means less value, when the opposite is true. Fewer features means each one gets more attention, more polish, more thought. Fewer features means a clearer identity. Fewer features means the user knows exactly what they're getting and why.
I've started applying this everywhere. Not just products. The portfolio site has nine project cards. Not twelve. Not twenty. Nine. Each one carefully chosen to tell a specific story about what I build and why. The blog doesn't cover everything I think about - it covers the things I think about enough to write well about. My daily reading is curated, not comprehensive. I follow fewer people and read them more carefully rather than following everyone and reading no one properly.
The edit as identity
Here's the thing nobody tells you about editing: it's an identity exercise. What you choose to include says something about what you value. What you choose to remove says something about what you won't compromise on. The edit is a statement. It says "this is what matters, and everything else doesn't." That requires knowing what matters. And knowing what matters requires knowing who you are.
When I look at the projects I've built - Modern Retro, Trove, CultureTerminal, First Out, the pub guide, all of them - the common thread isn't the technology or the category or the audience. The common thread is curation. Every single one of them is, at its heart, an editing exercise. Taking something vast and unwieldy and reducing it to something focused and useful and, hopefully, beautiful. A culture feed that shows you what matters, not everything that exists. A tube exit guide that tells you where to stand, not everything about the tube. A pub guide that recommends the right pub, not every pub.
That is taste. Not the ability to recognise quality - everyone can do that. The ability to choose. To commit. To say "this, and not that." To look at forty ideas and build eleven. To look at a feature list and cross out half of it. To look at a menu and serve four dishes instead of forty.
The hardest part of building anything isn't the building. It is the editing. It's looking at something you've made and asking "does this need to be here?" with genuine honesty. It's killing features you spent time on because they dilute the whole. It's saying no to good ideas because you already have a great one.
Anyone can add. Taste is knowing what to take away.
