The Birth of the Web: A Love Story Between Documents

The Birth of the Web: A Love Story Between Documents

The Library at the End of the World

In the late autumn of 1988, somewhere in the suburbs of Geneva, Switzerland, a young British scientist sat at his desk in a converted farmhouse on the edge of CERN, the European Organization for Nuclear Research. Outside his window, the Swiss Alps stood silent against the winter sky, their peaks dusted with early snow. Inside, Tim Berners-Lee stared at a problem that had been nagging him for months—a problem that seemed, on the surface, laughably mundane compared to the particle physics experiments happening in the massive underground accelerators just a few kilometers away.

CERN was, by any measure, one of the most remarkable places on Earth. It was where scientists had discovered the W and Z bosons, where the antiproton decelerator hummed with activity, where some of the brightest minds on the planet collaborated on questions about the fundamental nature of reality. Yet for all its scientific grandeur, CERN suffered from a decidedly unglamorous affliction: it could not keep track of its own information.

Tim's job title was "Software Engineer" in the computing services division, but what he really did was help scientists manage the chaos of their research. And chaos it was. Thousands of researchers flowed through CERN every year, each working on different projects, generating mountains of data, writing countless reports, and developing software that was often as unique as the people who wrote it. The problem wasn't that this information didn't exist—it existed in abundance. The problem was that it was scattered across dozens of incompatible computer systems, locked in proprietary databases, hidden in forgotten directories, and buried in email archives that no one could search. Finding a specific piece of information was like trying to find a single book in a library where every book was written in a different language and shelved by a different librarian using their own private system.

This wasn't a new problem. Libraries had existed for thousands of years, and humanity had developed sophisticated systems for organizing knowledge: the Dewey Decimal System, the Library of Congress classification, card catalogs, and indices. But those systems assumed that someone central—a librarian—would maintain the organization. What happened when there was no librarian? What happened when everyone was both a reader and a writer, when new documents appeared constantly, when no single authority could dictate how things should be organized?

Tim picked up a pen and began to write. What emerged over the following months would become one of the most influential technical documents in human history: a proposal titled "Information Management: A Proposal." Submitted in March 1989, it contained the seeds of an idea that would fundamentally transform how humanity stores, retrieves, and shares information.

The Vannevar Bush Vision

To understand what Tim was trying to accomplish, we need to travel back in time—not quite as far as the Library of Alexandria, but close. In 1945, as World War II was drawing to a close, a brilliant American engineer and science administrator named Vannevar Bush published an essay in The Atlantic Monthly that imagined a device he called the "memex." Bush, who had served as an advisor to President Roosevelt and overseen the wartime coordination of scientists, was worried about a different kind of information crisis: the explosive growth of scientific knowledge was making it impossible for any single researcher to keep up with developments in their own field, let alone adjacent fields.

The memex, as Bush imagined it, would be a desk with screens and levers, capable of storing all of a person's books, records, and communications. But more than a simple storage device, it would allow its user to create "trails" through the accumulated knowledge of humanity—connecting related concepts, jumping from one document to another, building personal pathways through the vast landscape of human understanding. "The human mind does not work that way," Bush acknowledged. "It operates by association. With one item in its grip, it snaps instantly to the next that is suggested by the association of thoughts, in accordance with some intricate web of trails carried by the cells of the brain."

Bush's insight was profound: human knowledge is not organized in neat hierarchical structures but in vast, interconnected webs of associations. Yet our information systems—libraries, indexes, databases—all forced us to impose artificial hierarchies on this naturally associative structure. What if, instead of forcing knowledge into boxes, we could follow the natural connections between ideas?

The memex remained a dream, but Bush's essay inspired others. In the 1960s, Ted Nelson, an American information theorist, began working on a system he called "hypertext"—a term he coined from "hyper," meaning beyond or above, and "text," meaning written words. Nelson envisioned a global system of linked documents, a "docuverse" where any piece of information could connect to any other, transcending the boundaries of individual documents, books, and databases. It was an ambitious vision, one that ran headlong into the technological limitations of the time. The internet existed in primitive form, personal computers were still years away, and there was no standard way to represent and link documents across different computer systems.

By the time Tim Berners-Lee sat down at his desk in Geneva, the concept of hypertext had been floating around in academic and technology circles for decades. Several experimental hypertext systems had been built, including one called HyperCard that Apple had introduced on the Macintosh in 1987. But these were all closed systems, islands of connection that couldn't communicate with each other. What was needed—what Tim would eventually create—was an open, universal hypertext system that could work across any computer network, connecting documents regardless of where they were stored or what kind of system they ran on.

The Proposal That Changed Everything

Tim's 1989 proposal was characteristically understated. He framed it not as a revolutionary breakthrough but as a practical solution to a management problem. "The current information systems have to be improved to accommodate the increasing complexity and amount of information that scientists and engineers produce," he wrote. But within the modest language lay an elegant vision.

At its heart, Tim's proposal described something deceptively simple: a system of "nodes" and "links." Nodes would be documents—any kind of document, whether a report, a database entry, a specification, or a piece of code. Links would be connections between these nodes, allowing a reader to jump from one document to another with a single action. This was the core concept of hypertext, but Tim added several crucial innovations that would make his system infinitely more powerful than anything that had come before.

First, the links would not be constrained by the document's internal structure. In most existing hypertext systems, a link was a property of a document—you could only link from one document to another if the system explicitly supported it. Tim proposed that links should be universal: any document could link to any other document, regardless of whether the target document was aware of the link. This meant that you could create a document about particle physics that linked to a document about the history of mathematics, even if the two were created by different people who had never met.

Second, the links would be bidirectional in concept but unidirectional in implementation. If document A linked to document B, you could navigate from A to B. But you could also, if you wished, navigate from B back to A—though B wouldn't necessarily have an explicit link pointing to A. This asymmetry was crucial because it meant that adding lin