There's a joke - one of those internet jokes that has been floating around for so long that nobody remembers who made it first - about reaching the last page of the internet. You've scrolled to the very bottom. You've seen everything. There is nothing left. Just a plain page that says: "Congratulations. You've reached the end of the internet. You can go outside now."
I think about this joke more than I should, because underneath the humour is a genuinely interesting question: what would be on the last page? Not the last page chronologically - not whatever was uploaded most recently. The last page in terms of significance. The final, definitive, most important thing the internet has to offer. If you could only save one page, what would it be?
This is a thought experiment about permanence. And permanence, it turns out, is something the internet is spectacularly bad at.
The disappearing web
We think of the internet as permanent. It isn't. Research suggests that a significant proportion of web pages from even a decade ago no longer exist. They weren't deleted in some dramatic act of censorship. They just... disappeared. The hosting expired. The company folded. The platform pivoted. The database corrupted. The domain lapsed. The internet isn't a library. It is a river. Content flows through it and most of it doesn't stay.
This troubles me for personal and professional reasons. Personally, because I'm a collector, a saver, a hoarder of links and references. My browser history is my diary. My bookmarks are my autobiography. And many of the links I saved five years ago now lead to 404 pages. The articles that shaped my thinking, the references I wanted to return to, the things I found important enough to save - gone. Not censored. Not removed. Just not maintained. Permanence on the internet requires active effort, and most things aren't important enough to anyone to justify that effort.
Professionally, this shapes how I think about what I build. Every product I make exists on rented infrastructure. The domain can expire. The hosting provider can change its terms. The free tier can become a paid tier. The technology can be deprecated. Nothing I've built is permanent in the way that a book is permanent. A book sitting on a shelf will outlast every website I've ever made, unless someone actively ensures those websites continue to exist.
What would survive
If we could only keep a fraction of the internet, what would we keep? This is a useful exercise because it forces you to think about value rather than volume. Not what's popular - what is important. Not what gets the most traffic - what would be missed most if it disappeared.
I think Wikipedia would make the cut. It's, genuinely, the closest thing the internet has produced to a public good. A free, comprehensive, continuously-updated encyclopaedia, written and maintained by volunteers, available in hundreds of languages, funded by donations. It's imperfect and it is extraordinary. If the internet had a last page, Wikipedia should be the second-to-last.
The Internet Archive would make the cut. It's, ironically, the thing that's trying to prevent the internet from having a last page at all - a comprehensive backup of the web, preserving pages that would otherwise vanish. The Wayback Machine is the internet's memory, and without it, the river metaphor would be even more alarming than it already is.
Personal websites would make the cut. Not all of them - but the ones that represent a person's authentic thinking over time. A blog that someone has maintained for a decade, where you can trace the evolution of their ideas, where the early posts are embarrassing and the recent posts are wise and the trajectory between them is the real content. These are irreplaceable. There's no algorithm that can reconstruct what a person thought about on a Tuesday afternoon in 2014.
Building for the long term
Most of the internet is built for the short term. Social media posts are designed to be consumed and forgotten. News articles have a shelf life of hours. Marketing content exists to drive a transaction and then becomes invisible. Even well-made products are built on technologies that become obsolete within a few years. The average lifespan of a website is measured in months, not decades.
I find myself increasingly drawn to the idea of building things that last. Not because I expect my pub guide to still be online in 2045 - I don't. But because the discipline of thinking about longevity changes how you build. If you ask "will this still be useful in ten years?" you make different decisions than if you ask "will this get traffic this week?"
The things that last on the internet tend to share certain qualities. They're simple. They're useful. They're well-made. They do one thing clearly rather than many things vaguely. They're built on stable technologies rather than the framework du jour. They have a point of view rather than a positioning statement. They feel human rather than corporate.
The last page is the first page
Here's where the thought experiment gets recursive. If the internet had a last page, what would it say? I think it would be something like this: "You made it here because you were looking for something. You always were. Every search, every click, every scroll was a question. The internet was never the answer. The internet was just a very large, very noisy place to ask."
The last page of the internet would be about the person reading it, not the internet itself. Because that's what all of this has always been about. We didn't build the internet to store information. We built it to connect humans. And connection isn't a technological problem - it's a human one. The best website in the world can't make you feel less alone if you don't know what you're looking for.
I think about this when I build products. Am I making something that helps someone find what they're looking for? Not in the Google search sense. In the deeper sense. Is this product helping someone understand their taste, organise their thoughts, discover something they didn't know they were interested in, connect with a place or a community or a feeling? If yes, it might be worth maintaining. If no, it might as well join the river.
What I want to survive
If I could choose which of my projects survives indefinitely, I would choose Modern Retro. Not because it's the most useful - it isn't. But because it represents something that only I could have made. The specific combination of cultural references, aesthetic sensibility, and creative vision that produced those images and that gallery exists in one person. If it disappeared, it couldn't be recreated, even with the same tools, because the taste that shaped it is unique.
That's, I think, the real answer to "what would be on the last page?" The things that should survive are the things that are uniquely human. The things that could only have been made by a specific person with a specific set of experiences and a specific way of seeing the world. AI can generate infinite content. Algorithms can organise infinite information. But only humans can create things that carry the fingerprint of an individual sensibility. Those are the things worth keeping.
So here's my conclusion, after this meditation on permanence and the internet and the pages that disappear when nobody is watching: build things that carry your fingerprint. Build things that could only have been made by you. Build things that aren't content but conviction. Those are the things that deserve to survive. Those are the things that someone, scrolling through the ruins of the web in fifty years, would stop and say: "a real person made this."
The last page of the internet isn't a page at all. It's the proof that someone was here, and that they made something worth remembering.
