From HTTP’s earliest request/response patterns to the first practical server interfaces, 1990–1994 quietly planted the design instincts behind today’s composable web APIs—uniform URLs, simple verbs, and the idea that you can stitch capabilities together through links and messages.
In the Web’s first years, “APIs” were often just URLs and forms hitting CGI scripts. Those primitive HTTP interfaces introduced a surprisingly modern idea: clients can influence what data they get back—an early ancestor of today’s client-driven data fetching and GraphQL-style thinking.
OpenAPI didn’t exist in the early 1990s, but the Web’s first standards—URLs, HTTP, MIME types, and simple server-side interfaces—introduced the core idea that an API can be described consistently. This chapter traces the roots of standardized API descriptions from 1990 to 1994.
In the early 1990s, the Web didn’t have RSS or Atom—but it did have the core ingredients of syndication APIs: stable URLs, machine-friendly representations, and repeatable retrieval via HTTP. This chapter traces how early HTTP interfaces and publishing habits made feeds inevitable.
A look back at 1990–1994, when the Web’s earliest HTTP patterns—resources, links, forms, and simple request/response behavior—quietly established the API ideas that later powered Facebook Platform APIs and the social graph.
Between 1990 and 1994, the Web’s earliest HTTP interactions weren’t called “APIs,” yet they introduced resource identity, uniform operations, and stateless requests—the same building blocks REST would later formalize.