There's a second-hand bookshop on Charing Cross Road that I visit whenever I'm in that part of London. I never go looking for anything specific. That's the point. I go in, turn left, start at whatever shelf catches my eye, and let the spines guide me. An hour later I'll walk out with two or three books I didn't know existed when I walked in. Books I never would have searched for, never would have been recommended by an algorithm, never would have appeared in any feed. Books I found by browsing.
You can't do that on the internet anymore. And that loss is bigger than most people realise.
The end of wandering
The early internet was built for browsing. The whole thing was a web -- you followed links from one page to another, each click a small discovery. Webrings connected sites about similar topics. Blogrolls introduced you to writers through other writers. You'd start on one page and end up somewhere entirely different, tracing a path that nobody planned and nobody could replicate. The browser was named for this behaviour. We browsed.
That's been systematically replaced by three things: feeds, recommendations, and search. Each one is useful. None of them allow for genuine serendipity.
A feed shows you what people you follow have shared. It's filtered, curated, and increasingly algorithmic. A recommendation engine shows you things similar to what you've already consumed. It's a mirror, not a window. A search engine shows you what you already know to look for. You need to form a question before you can get an answer. None of these replicate the experience of wandering into a space and finding something you didn't know you wanted.
What browsing actually does
Browsing isn't aimless. That's the misconception. People who browse -- whether in bookshops, record stores, charity shops, or the old internet -- aren't wasting time. They're exposing themselves to the unexpected. And the unexpected is where taste comes from.
Think about how you developed your interests. Not the obvious ones -- the ones that surprised you. The band you discovered because their album was filed next to something you were looking for. The book you picked up because the cover was interesting. The website you landed on because someone linked to it from somewhere else entirely. These accidental discoveries form the edges of your taste. The centre might be predictable, but the edges are where things get interesting.
Algorithms don't do edges. They do centres. They learn what you like and give you more of it. That's useful when you know what you want. It's devastating when you don't know what you don't know. The algorithm can't show you the thing you'd love but would never search for. It can only show you variations of things you've already found.
Physical browsing as resistance
I've become almost evangelical about physical browsing. Bookshops, record shops, charity shops, markets. Places where things are arranged in space rather than ranked by relevance. Where you encounter objects by proximity rather than by preference.
A charity shop is the purest form of browsing that exists. There's no algorithm. No recommendation engine. No search function. Just a chaotic collection of things that other people no longer wanted, arranged on shelves by people who may or may not have a system. You rummage. You discover. You find a first edition of something brilliant next to a dog-eared Jackie Collins. The juxtaposition is the experience.
Record shops work the same way. You flip through the crates. You see album covers that interest you for reasons you can't articulate. You pick something up, put it back, pick up something else. The physical act of handling objects creates a different relationship with discovery than scrolling through a screen. You commit more to each encounter. You slow down.
I take my kid to bookshops for this reason. Not because I'm anti-screen or pro-paper in some performative way, but because I want them to know what it feels like to discover something by accident. To understand that the best things aren't always the things you went looking for. That serendipity is a skill you can develop, not just luck you hope for.
The internet we lost
I'm nostalgic for a version of the internet that may never come back. The internet of personal homepages, of "cool links" pages, of webrings and web directories and the Yahoo homepage when it was actually curated by humans. The internet where you could click a link, end up somewhere strange, and think: I had no idea this existed.
RSS was the last great browsing technology. You subscribed to feeds from people and publications you found interesting, and you read them in order, without ranking. No algorithm decided what was important. You just scrolled through and your own taste did the filtering. It was browsing in digital form -- exposure to a wide stream of things, with your brain doing the curation.
I still use RSS. It makes me feel slightly like a person who insists on writing letters in the age of email. But the experience is genuinely different from any social media feed. I encounter things I wouldn't have chosen. I read entire pieces instead of scanning headlines. I follow trains of thought instead of being jolted between unrelated fragments. It's slower and that slowness is the feature.
Building for serendipity
This is why I keep building curation tools. Trove, CultureTerminal, even the pub guide -- they're all attempts to recreate some version of the browsing experience online. Not algorithmically. Not by recommendation engine. But by putting interesting things in proximity to each other and trusting the user to find their own path through them.
The design challenge is real. How do you create digital serendipity without creating chaos? How do you help people find things they didn't know they wanted without overwhelming them with things they don't want at all? Physical spaces solve this with spatial design -- you walk through the shop and the layout guides your attention. Digital spaces have struggled to replicate this because screens are flat and feeds are linear.
But I think there's an answer in the idea of adjacency. Not "people who liked this also liked that" -- that's recommendation, not browsing. Real adjacency. Things placed next to each other not because they're similar but because they're interesting in combination. A link about Japanese design next to a link about barbershop culture next to a link about the economics of independent bookshops. No obvious connection. But your brain starts making connections anyway. That's browsing.
A small act of defiance
I walked into that Charing Cross Road bookshop last month and bought a book about the history of department stores. I wasn't looking for it. I don't have a particular interest in department stores. But the cover caught my eye, I flipped through it, and something about the photography and the writing pulled me in. It turned out to be one of the most interesting things I've read this year -- full of ideas about retail as theatre, about how physical spaces create desire, about the lost art of the shop window as cultural statement.
No algorithm would have shown me that book. No recommendation engine would have connected it to my interests. No search query would have led me to it. I found it because I was browsing. Because I walked into a physical space with no agenda and let the shelves do the work.
That's not just a pleasant experience. In a world that's increasingly filtered, ranked, and personalised, it's a small act of defiance. Against the algorithm. Against the feed. Against the tyranny of relevance.
Sometimes the most relevant thing is the thing nobody thought to show you.
