“The people is one, and they have all one language; and this they begin to do: and now nothing will be restrained from them, which they have imagined to do.” The Book of Genesis

“The Internet is becoming the town square for the global village of tomorrow.” Bill Gates, Business @ the Speed of Thought

“There are downsides to everything; there are unintended consequences to everything. The most corrosive piece of technology that I've ever seen is called television -- but then, again, television, at its best, is magnificent.” Steve Jobs, Rolling Stone, 2003

There were 60 websites in 1992. 623 websites in 1993—including Bloomberg, Wired, and IMDB. It was possible to read everything on the internet. But then, there were 2,738 sites by mid-1994, over 10,000 by the end of the year. You’d have to read 27 full websites a day to keep up.

Yahoo! launched that year, as a human curated directory of everything worth reading on the internet. A mere four years later, AI neé machine learning took over, with Google using backlinks and the reputation they inferred to tame the internet.

WordPress came in 2003. Facebook the next year, Reddit the year after, Twitter the next. For a while, again, you could read every post from your favorite blogs, scroll through all of your friend’s social media posts. Then it became too much, and algorithmic feeds came to the rescue, only to isolate us into filter bubbles. Then came AI generated content. Then came everything.

You can’t keep up with it all today. Nobody can.

We thought we’d get an explosion of creativity, of expression, of freedom, if everyone joined the internet. And we did, with shared memes replacing the shared moments of seemingly everyone watching the same TV shows in decades past. With lovingly crafted sites you’d stumble upon, dreams and dreamers, and realize that someone else in the world had the same niche interests as you. That wasn’t so bad.

And then we completed the internet. Google’s full of answers to seemingly everything; YouTube’s optimized for retina-searing thumbnails; TikTok’s QVC for a new generation; Twitter went from what you had for lunch to what everyone’s angry about today.

Everything’s done; time to pack it up and fulfill that perennial tech-adjacent dream of starting a bookstore or café. 

Maybe everybody is too many.

So you decide to open that bookstore or café. You could open it anywhere. A café in the middle of nowhere has no competition—and no audience. A bookstore in a mall like my local one might compete with 4 other shops—and be noticed by thousands every day. 

Start a coffee shop, and you’ll need to convince people to choose your coffee over 5 existing cafés—along with restaurants that serve cups of joe—but hey, at a higher percentage of those mall visitors will buy coffee than books on any given day.

Want to open a website writing about Apple? 13,200,000,000 other websites already did.

And, sure, some of those are talking about the fruit, others are talking about the Beatles’ company. The vast majority, though, are talking about Steve Jobs’ tech company, competing with your new site for readers.

Even if you focus on a niche, say on purple iPhone 15 Pro Max leather cases, Google returns 17 million results. Niche way down, write about iPhone Error Code 7005, and Google still shows 57 results.

You’ve got a better shot at opening a coffee shop and convincing your fellow shopper to buy your coffee over Starbucks than you have convincing your fellow web shopper to visit your website.

All because in the real world, location restricts competition.

There are over 38,000 Starbucks in the world. But your theoretical coffee shop isn’t competing with all of them. It’s competing with those in perhaps a 15 minute walking distance of your store. Sizable competition in a dense city, perhaps, where every other shop seems to serve coffee, but far less competition than online. Sell coffee beans on your website, and you’re just one of 4.7 billion websites mentioning coffee. Sell it on Amazon, and your coffee’s one of 20,000 options.

And it’s not like you can just open a coffee shop on a whim. The retail location will need a vacancy, and they’ll need to want your shop enough to agree to let you lease it. Then you’ll need to pay them for that privilege, pay them perhaps even a percentage of your total sales. You’ll need grinders and espresso machines, takeaway cups and trash cans, suppliers and staff. You’ll spend six figures before making a dime. So few people do it. The entry costs are too high.

The internet, with its nearly zero marginal costs, is wide open. Anyone can publish anything on a whim, start selling coffee for the cost of a cup of coffee—or less. Marc Lore and Vinit Bharara started Diapers.com—later to be sold to Amazon for $545 million—by literally buying diapers at local wholesale stores—no supplier contracts or warehouse leases or investment required beyond a domain name and a bit of code and design and a Costco membership. Do that in the real world, and at the very least you’re going to need a storefront lease, shelving, staff, and a dozen other things that would make you think twice before starting a diaper store.

Getting content on the internet, if anything, has gotten cheaper. Diapers.com needed a server—or expensive hosting, at least. Anyone today can start a Shopify store for $20, write on a Medium blog for free, send Substack emails to 100k people without paying the $400+ that most traditional email platforms charge, or host videos on YouTube that get a million views without paying anything for bandwidth.

It’s getting noticed that has gotten more expensive. You can do more for free than before, but you’ll get fewer results without paying. Grabbing one of those early nearly-free plots of digital land almost guaranteed visitors originally, something repeated with each new major platform. Then as they fill up, free infinite space starts looking rather expensive as the longtail gets ever longer. What isn't rare isn't valuable, and there's never been inflation as exponential as there is in today's digital property market.

Because imagine competing with those 5 other cafés in a mall. You’ll get some business just from people walking buy, more with better coffee, or cheaper, or more unique offerings, or a combination of all three. Or, you might opt to buy ads beside escalators to get attention, perhaps, or pay more in rent for a location near the entrance.

On the internet, that means being more prolific, to get the algorithms to send you traffic. It means paying for search and social ads, to show up in the results pages and feeds that formerly gave you free traffic. It means doing something unique, crazy even, to stand out. Just opening your virtual doors will do nothing. “In a future where AI makes products cheap, investing in distribution will be the wisest bet,” wrote Evan Armstrong recently, something that’s already come true for content as the free distribution routes have filled up with cheap content, generated by AI and humans alike.

The platforms needed your creations when they were new. They’d send you traffic for free, subsidized by investor cash to grow fast. But now they’re the entrenched players, as important to our 401k’s as GE was to a previous generation’s. Thus the cycle that Cory Doctorow calls the “enshittification” of everything: You make something free and better for people to attract an audience, then you make it better for businesses at the expense of users to make money, then you make it better at making money than for the businesses to make more money, then the users start leaving an increasingly poor experience, then everything dies.

Maybe it all won’t die. But search and social aren’t a great place to build a new business, capture a new audience, in 2024.

And maybe, just maybe, “enshittification” isn’t entirely a problem of greed. Maybe the true problem is that we put everything and everyone in the same place, and the world’s a bigger place than a single internet. If so, scale’s the problem, and smaller’s better.

Back to the website

You don’t need the whole internet. “The long tail is famously good news for two classes of people; a few lucky aggregators, such as Amazon and Netflix, and 6 billion consumers,” wrote former Wired editor Kevin Kelly. “But ... a decidedly mixed blessing for creators.” The solution was what Kelly called 1,000 true fans. Have those people who love your work, and the internet, for you, is like the early ’90’s. People will come directly to your website to use your app, read and watch what you create, and interact with other fans.

Now how to get those fans, without relying on search and social and increasingly crowded platforms?

There’s no single solution, but there are hints at what works from around the internet.

Take Facebook. The blue app—Facebook’s original social network—is, for many of us, increasingly filled with irrelevant content. There are only so many long-lost high school classmates to reconnect with. Maybe all is well—and in developing regions where Facebook adoption has been slower, where fewer businesses have standalone websites, Facebook is still growing—but it feels to most like engagement’s dropped off.

But Facebook Groups? They’ve exploded, from 100 million people using Facebook Groups in 2017 to 400 million in 2019 then 1.8 billion in 2021.

Single-topic groups work. You’re not interacting with all of Facebook. It’s not the whole world. You’re, instead, in this smaller convention center, of sorts, chatting with other people about this one topic. Maybe you have nothing in the world in common with those other people; maybe you’d debate them on different issues, or even dislike them in real life. But here, in this group, you’re all talking about succulents or skydiving, you’re meeting new people and gaining new ideas about that topic, and it’s the best of the internet for you.

Smaller, old-school forums serve the same niche. They’re popular, with fiercely devoted fans—as Amazon found out when they decided to shut down the camera forum Dpreview only to reverse course and sell it instead. Communities like Reddit have tapped into the same vein, with people excitedly participating in a handful of communities they like even if they seldom browse the larger network.

The good of single-topic groups is that they’re focused. Great for discovering like-minded people, about that one thing at least. Great for connecting with your fans, if all they’re talking about is your products. Less good, though, about getting a new thing discovered, at capturing those first thousand fans.

Single demographic forums—like Hacker News, which is focused on “Anything that good hackers would find interesting,”—are even more promising. They also seem harder to pull off, seem rarer and more likely to buckle under the same weights pulling the rest of the internet apart. But the trick of Hacker News, the thing that keeps you going back, is that it covers such a wide range of interests. Domain names. Tech history. AI. Work and life balance. Coding tricks. A new app that does something unique. You never know what you’ll find, exactly, but if you share similar interests to its reader base, you’ll always find something that strikes your interest.

Magazines served this function—and still do, to a degree. You’d like the general ethos of Wired or New Yorker or National Geographic, and pick them up not because you knew exactly what you’d read, but knew the general vein of topics they’d cover. You’d find something that goes more in-depth about something you knew, discover something else you’d never heard of, all because it was put together in the same place. If anything, the gradual “enshitification” of the broader web and social in general may, as Ben Thompson recently put it, “by extension [make] destination sites that much more valuable, which is to say it makes the New York Times more valuable.” The same applies for other sites that are as curated and focused as a high-quality magazine.

The tough part is staying on topic, of keeping the community from devolving into ads and flame wars. It’s like how Scott Alexander phrased virality and compounding effects: “we’re looking at an exponential process, like R for a pandemic. If the exponent is > 1, it gets very big very quickly. If the exponent is < 1, it fizzles out.” If you can get your community with R > 1, it very quickly becomes self-sustaining, and the good people who inevitably become volunteer moderators stave off the trolls (and if not, your community will quickly devolve into an R < 1 toxic group). 

With luck and self-sustaining growth, though, a community can become a discovery engine to rival search and social. And if you're a frequent and genuine participant in these communities, it's almost always easier to determine what its readers will like and reward, a thesis that will resonate with them, than to guess what Google might rank on the first page.

Email newsletters—or, again, self-selected single-topic communities—are, perhaps, the bridge between those broader interest communities and your true fans. You filtered the whole internet (too many people) down to people who like things similar to your thing (still a lot, but manageable—call it a trade show, perhaps), to people who really, really like what you made (a few, but just enough—your small but viable coffee shop). You want to grow the bottom of that funnel and keep all of them thinking about you from time to time. Thus email newsletters, the one notification you can push without an app, the one feed that’s still for the most part chronological, the one cross-account communication system that still works.

And then, share to social; can’t hurt. Test out new platforms; maybe a new one will come along and you can snag a bit of early adopter free real estate. Don’t hoard what you create; no risk in sharing your article on multiple blog platforms, or putting your videos on YouTube and TikTok and more, as you never know what will work. 

Follow generally good practices for your website, and let search take care of itself; one of the best benefits of Hacker News or other forums is if people like what you wrote and share it, you’ll get backlinks without trying and start ranking for keywords you never would have thought of. Take your online offline and do meetups and events; make them happen, if they haven’t already, and maybe you can snag some virtual attention in real life.

So communities. People. Single-topic communities with fans of one thing. Shared-interest communities with people who have common interests. Sites and newsletters and other ways to connect directly with people who are already interested so you don’t have to worry about Google and social and re-discovery.

Is that the future? Maybe. Is there something better that’s possible? Maybe.

But as much as tech is interesting for its own sake, as much as it’s made our current lives possible, it’s people who are actually interesting. And limitations. You meet people at a slow, local coffee shop or a conference in a way you never would in an airport or mall. We love an open-world game, the freedom to do anything, but maybe we need the boundaries of a path to guide us, a bit.

Image Credit: Header photo by Owen Cannon via Unsplash.