Where Exactly Is the Internet?

When you "go online," where are you going? When you "visit a webpage," what are you visiting? When you "enter a chat," where have you entered?

The obvious answer is “nowhere”. Beyond that it’s technical. The Internet is in fiber-optic cables, data centers, routers, satellites, servers, wireless signals, and the device in your hand. A webpage lives on a server. Your browser sends a request, data comes back, and the page is rendered. This answer is correct, but it only explains the infrastructure, not the experience.

We go online. We browse and visit sites. We follow links. We gather in forums and lurk in threads. This language is usually treated as metaphor but it isn't just that. It records the shape of the experience. You sit in a seat and peer through the browser, you feel yourself transported elsewhere. Surfing the web, scrolling your Twitter feed, getting lost in Reddit threads — time dissolves as you move through and engage with this other dimension. For a time, the room you're sitting in quietly disappears. This experience has an origin story.

ARPANET connected its first nodes in 1969. By the 1980s there was a functioning global network carrying email, file transfers, remote sessions, and discussion forums. Usenet, FTP, Telnet, Gopher — these were working systems, used daily for collaboration and research. The Internet was not waiting to be born in the 90s, it was already there.

But it remained the domain of academics and technical hobbyists.

It worked fine, but navigating it required initiation. You had to know the address of the machine you wanted to reach, the protocol to speak to it, the commands to type into a terminal, and how to interpret what came back as raw text. The network was powerful, but sparse and dark. It was like traveling between cabins in deep forest — isolated clearings, connected by unmarked trails, accessible only to those who knew where to look and how to speak the language of the frontier.

Then, in the early 1990s, a new layer appeared on top of the network: the World Wide Web. A handful of interlocking innovations changed not what the Internet could do, but what it felt like to use it. The web gave information a visible surface. A document could now have layout, images, and embedded links to other documents. URLs gave locations stable, shareable, memorable names. You could point someone to a place and they could go there. The hyperlinks connected sites into a traversable landscape — you could navigate cyberspace by choosing a path. And the browser gave you a window to look through. The verb itself tells you what changed; you could now browse. Regular people could look around.

The explosion that followed is usually told as a story about commerce. Companies went online. Money moved. Fortunes were made and lost. But the deeper story is about cognition. A global communications network became publicly habitable practically overnight. Hundreds of millions of people who could never have navigated the old net walked into the web in the span of a few years, because it was presented in a familiar language: pages of information, sites with addresses to visit, links as doorways between one place and the next, rooms to occupy and converse in, going somewhere and coming back.

Why did this work?

There is an ancient mnemonic technique called the memory palace. To remember a sequence of ideas, you imagine a familiar building and place each item inside it — one image by the door, another in the hallway, another on the table, another at the foot of the stairs. Then you walk through the imagined space and retrieve what you stored there.

The technique works because spatial cognition is old and deep. We remember by locating, understand by arranging. We make the abstract graspable by giving it position, distance, adjacency, containment, path. This capacity is fundamental architecture, intrinsic to how we navigate and make sense of reality.

The web didn't invent the sense of navigable, non-physical place — it exploited a capacity that was already there.

The structure of the web was informed by an impulse to find symbiosis between contemporary distributed computation and existing cognitive structures. Its rapid, global adoption is the result of that convergence finding purchase. The pre-web Internet proves this by contrast. Same technical substrate. No spatial interface. No mass adoption. What was missing was not capability but a format a mind could project into.

There is no single Internet. Granted, there is one shared infrastructure — protocols, cables, servers, platforms, devices, laws, companies, power grids. But there is no single experienced web. Your Internet is not mine. A teenager's Internet is not a retiree's Internet. People in Oakland, Lagos, Seoul, and rural Montana do not enter the same space. They don't even enter the same space as their neighbors or friends. They may touch the same platforms, pass through some of the same systems, encounter the same viral fragments. But the lived territory differs, because each mind assembles it from different memories, habits, languages, communities, and histories.

This is also true of physical places. A city has public streets, but each resident carries their own private map — the corner where something happened, the block never walked after dark, the café that means nothing to one person and everything to another. The physical world provides the signals, but the place we actually experience is assembled in the mind. The web lays this bare.

So where exactly is the Internet?

It is in the machines, but that is only the infrastructure, the storage and delivery apparatus. The 'place' of the Internet is assembled in cognition, because the web was shaped, over decades, to be assembled that way. It is a planetary information system that became universally navigable the moment it learned to suit the oldest parts of the mind.

You "go online" because the machinery evolved to become a recognizable somewhere.