South Korean girl group KiiiKiii has released a new track titled 404 (New Era), a single from their second extended play, Delulu Pack. The song takes its name from one of the Internet’s most familiar (and frustrating) messages: the HTTP 404 error. Rather than treating it as a dead end, the group reframes it as “a symbol of freedom without coordinates”. So what exactly is a 404 error, and why does it resonate far beyond web browsers?

Over time, the 404 error has evolved into shorthand for absence — something missing, unavailable, or momentarily out of reach. In Internet culture, it has even become a meme, sometimes used to describe a lack of presence of mind or emotional disconnect. Along the way, a popular story emerged to explain its origins.

According to this widely shared myth, the 404 error was named after Room 404 at CERN, which supposedly housed the world’s first web servers. The story goes that Tim Berners-Lee, inventor of the World Wide Web, had his office there and was often “not found”.

It’s a neat story but completely untrue. That claim was dismissed by Robert Cailliau, a pioneer of hypertext systems that helped shape the modern web. Speaking to Wired in 2017, Cailliau said: “404 was never linked to any room or any physical place at CERN.… That’s a complete myth.”

To understand what a 404 error really means, it helps to look at how the Internet works at a basic level.

The web is built around two core components: clients and servers. The client is usually your web browser — Chrome, Safari, or any other — which sends a request when you type in a web address. That request is sent to a server, which then responds with the data needed to load the page.

This exchange relies on HTTP, or Hypertext Transfer Protocol. Protocols are simply agreed standards that allow computers across the Internet to communicate with one another. Every time a client makes a request, the server replies with a status code that explains what happened.

These status codes are grouped into blocks. Codes in the 100 range are informational, the 200 range indicates success, the 300 range is used for redirection, the 400 range signals client-side errors, and the 500 range points to server-side problems.

The 404 error sits squarely in the 400 block. In plain terms, it means the requested page could not be found. Other codes in this group include 400 for a bad request, 401 for unauthorised access, 402 for payment request, 403 for forbidden pages, 405 for method not allowed, 406 for not acceptable, 407 for proxy authentication method and 408 for request timeouts. The list carries on

Despite its reputation, a 404 error is not always a sign that something has gone wrong. If a user mistypes a URL or follows an outdated link, returning a 404 is the correct response. There’s a persistent belief among website owners that having any 404 pages is inherently bad. That isn’t true.

If a well-established page with strong link equity suddenly returns a 404, it may make sense to redirect users to a relevant alternative. But for a genuinely non-existent page, leaving it as a 404 is often the right choice.

In that light, KiiiKiii’s reimagining of the error feels apt. A 404 isn’t always a failure… sometimes it’s simply a sign that you’ve stepped off the mapped route, into something undefined, unlabelled, and unexpectedly free.