Forgotten Foreman: The URL Architect of the 90s Academic Web

Before 'Information Architecture' was a job title, and long before canonical tags or redirect maps were a whisper in a developer’s ear, there was a different kind of architect. They didn’t work with blueprints, but with directory trees on university servers. They weren’t designing for Google’s crawlers, but for the handful of human colleagues who might navigate a Gopher hole or an early HTTP site. Their tool was a consistent naming logic, and their legacy is buried in the sunken continents of the early web.

I’m thinking of a specific figure, though they stand for many: the departmental sysadmin or graduate student in the mid-1990s tasked with “putting the lab online.” They were handed a jumble of papers: syllabi, CVs, pre-print articles, technical reports, and a dozen scanned images from a lab experiment. Their canvas was a public_html folder, often with a strict quota. Every decision they made—where to place a file, what to call it, how to link from the index.html “hub”—was a foundational act of URL design.

The Invisible Logic of Necessity

Their structure wasn't born from abstract theory. It was forged in the fire of immediate necessity and foresight for future updates. A folder named /papers/1996/ wasn't just organization; it was a promise. It said: next year, you will add a /1997/ folder here, and the old links won’t break. A file named smith_cv_1995.html anticipated smith_cv_1996.html. This was manual, deliberate versioning. The internal links they crafted in their hand-coded HTML were the site’s entire circulatory system. There was no search box. Navigation was a tree you had to climb correctly.

This foreman’s most profound, and most tragic, contribution was the redirect—though they’d never call it that. When Professor Smith finally got a professional website and the old /~smith/ directory had to be deprecated, the solution was often a single, plaintext index.html file left in the old location. It contained one line of HTML: a meta refresh tag or a simple “This page has moved to…” with a hyperlink. This was a redirect born of courtesy, a breadcrumb left in the digital woods so the few who had bookmarked it wouldn’t hit a 404 and think the knowledge was lost forever.

We’ve automated and systematized all this now. Our CMS generates hierarchies, our plugins manage redirect chains, and our canonical tags resolve duplication we can't even see. But something of that tactile, human-scale responsibility has faded. The early web’s URL architect worked with a palpable sense of place. A URL was a genuine address, a path to a specific file in a specific folder on a machine they could point to in a server room. Breaking a link felt like moving a building and leaving no forwarding address.

We navigate the sleek, dynamic web they helped dream into being. But occasionally, when you stumble upon a perfectly preserved academic site from 1997, with its logical folders and that last, considerate “This site has moved” page, you’re seeing the handiwork of that forgotten foreman. They laid the first bricks, not for algorithms, but for people, establishing a quiet ethic of structure and permanence that we’re still trying to live up to.

Notes & further reading

A few pages I came back to while writing this: