Ancient Web: A 2001 Interview Saved the Story of America’s First Web Server

Part of the Ancient Web Collection.

The first American Web server was not installed to sell anything. It was installed because a physicist wanted his colleagues to search a database without having to learn another computer's commands.

Read the surviving Paul Kunz interview.

This page preserves Paul Festa's December 10, 2001 CNET interview with Paul Kunz of the Stanford Linear Accelerator Center, or SLAC. It is a reprint of a contemporary conversation, not the original 1991 web page itself. That difference matters: the valuable artifact here is Kunz's own explanation of what happened.

Kunz remembered visiting CERN in Geneva on September 13, 1991, when Tim Berners-Lee showed him the WorldWideWeb browser running on a NeXT computer. The demonstration was not just about reading linked text. It could search a remote system, and that suggested an immediate use for SLAC's SPIRES database of high-energy-physics literature.

A useful application came before mass adoption

Kunz says the SPIRES database already contained roughly 200,000 paper references. Physicists could reach the information by other methods, but commands and remote logins were cumbersome. He returned to California intending to put a friendly Web interface in front of it. Other priorities delayed the job until December 1991.

The interview remembers a wonderfully primitive transatlantic test: browser software ran on a computer at SLAC while its display was sent back to a screen at CERN. Once the server was operating, a user could search scientific preprints and look up telephone-directory entries.

Kunz also describes a January 1992 physics workshop in France where Berners-Lee demonstrated access to SLAC's database. The audience already knew the underlying reference collection. Seeing it become accessible through links, Kunz recalls, helped them imagine many more uses.

His discussion of line-mode browsers is another small treasure. Without a NeXT machine, users followed numbered links by typing numbers rather than clicking anything. Early Web adoption was not necessarily a graphical experience.

The interview includes later recollections about Tony Johnson's dynamic pages and the growing need for graphic displays of scientific papers. Some claims of technological priority are Kunz's recollections and deserve that attribution.

Here, on an ordinary HTML reprint, the supposedly abstract birth of the American Web becomes a practical story about making a physics library easier to use.

Read the full 2001 interview.

Part of the Ancient Web Collection.