The Web Is Quietly Forgetting Itself

This site is, in a small way, a survivor. It went dark for years and was pulled back from the archives of the early Internet, which gives me an odd vantage point on a problem most people never notice until it bites them: the web forgets. Not dramatically, not all at once, but steadily, in the background, as links break and pages vanish and whole corners of the network simply stop answering. We tend to think of the Internet as a place where nothing is ever truly lost. The truth is closer to the opposite. The web is one of the most impermanent things humans have ever built, and it is quietly forgetting itself.

The quiet epidemic of dead links

Anyone who has followed a link into a "404 Not Found" knows the small thud of it. What is easy to miss is how common it has become. Links rot constantly — a page moves, a site shuts down, a company folds, a domain lapses, and every link that ever pointed there turns into a dead end. This is not a rare accident; it is the normal condition of the web over time. A link that worked five years ago has a real chance of being broken today, and the older the link, the worse the odds. The phenomenon even has a name that has been around for a long while: link rot.

The effect compounds because the web is built entirely out of links. It is not a library of self-contained books; it is a dense web of pointers, each page referring to others. When those pointers break, the structure frays. A researcher cites a source that later disappears. An article links to the evidence for its claim, and the evidence evaporates, leaving an assertion with nothing behind it. Follow enough old links and you begin to feel the web's memory as something full of holes — a record that is constantly deleting its own footnotes. We built a civilisation's worth of knowledge on a foundation of addresses that were never guaranteed to keep pointing anywhere.

Why the network decays

It helps to understand that this decay is not a bug someone forgot to fix; it follows from how the web is made. A web page is not a durable object like a printed page. It is a file sitting on a computer somewhere, served only as long as someone keeps that computer running, keeps paying for the domain, and chooses to keep the page online. The moment any of those conditions lapses — the bill goes unpaid, the company dies, the owner loses interest — the page stops existing, and there is no central authority that preserves it. Persistence on the web is an active effort, not a default. Things stay up only because someone keeps holding them up.

There is also a cultural half to the problem. The modern web has drifted toward the ephemeral almost as a matter of taste. So much of what people produce now lives inside platforms and feeds designed for the moment — content that scrolls past and is gone, with no stable address at all. The old web, for all its clumsiness, was built out of pages: things with locations you could bookmark and return to. The engineered impermanence of the feed is a cousin of link rot, a different way the network refuses to hold still, and one I have grumbled about before in the feed never ends. Between technical fragility and a culture of the disposable, forgetting is the path of least resistance.

What we lose when a page dies

It is tempting to shrug at this. Pages die, new ones appear, the web is bigger than ever — so what is actually lost? Quite a lot, it turns out. The web is not just entertainment; it has become the working memory of a great deal of human activity. Public records, research, journalism, technical documentation, personal writing, the accumulated explanations that let people learn things — an enormous share of it now exists primarily, or only, as web pages. When those pages rot, we are not losing noise. We are losing the citations under arguments, the evidence behind claims, the record of what was said and known at a given time.

The subtler loss is to history itself. Future historians will try to understand our era largely through what we published online, and they will find a strange, patchy archive full of gaps exactly where the most ordinary and everyday things used to be. The blockbuster events may survive; the texture of daily life, the small sites and personal pages, will often be simply gone. A civilisation that records almost everything and preserves almost nothing is a peculiar thing to be, and we have become it without deciding to. The web's forgetting is not neutral. It quietly edits the past in favour of whatever happened to stay online.

Against the forgetting

I do not want to end on pure lament, because the forgetting is not total and it is not hopeless. There are people and institutions who take the problem seriously — the archivists, formal and informal, who work to capture pages before they vanish, so that a dead link does not always mean a lost page. Their effort is one of the quietly heroic projects of the Internet age, and much of the early web that still survives at all survives because of it. This very site is a small piece of evidence that things thought lost can sometimes be brought back, if someone cares enough to do it.

Individually, the lesson is humble but real: if something on the web matters to you, do not assume it will be there tomorrow. Save it. Keep your own copy. Treat the pages you rely on as mortal, because they are. And if you make things for the web, consider giving them stable homes and keeping them up, resisting the drift toward the disposable. The old web worked because people maintained their corners of it by hand and expected them to last. That expectation is worth reviving. The network will keep forgetting on its own; remembering is the part that takes deliberate, stubborn effort — and it is, I think, one of the more worthwhile things a person can spend that effort on.