The Road to Xanadu:
Public and Private Pathways
on the History Web

Roy Rosenzweig

On August 24, 1965, Theodor Nelson presented a paper to the Association for Computing Machinery national conference in which he introduced the word "hypertext" to refer to "a body of written or pictorial material interconnected in such a complex way that it could not conveniently be presented or represented on paper." Nelson, who had started musing about this sort of associative thinking and linking as a Harvard University graduate student in 1960, viewed "hypertext" as an integral part of an imagined globally interconnected library and publishing system that would "grow indefinitely, gradually including more and more of the world's written knowledge" and "have every feature a novelist or absent-minded professor could want, holding everything he wanted in just the complicated way he wanted it held, and handling notes and manuscripts in as subtle and complex ways as he wanted them handled."1	1
Two years later, while working at the publisher Harcourt Brace, Nelson—an inveterate coiner of terms whose own Web page lists sixteen words or phrases that he claims to have introduced into general use—started to describe his global library as "Xanadu." "For forty years," Nelson wrote recently, "Project Xanadu has had as its purpose to build a deep-reach electronic literary system for worldwide use and a differently-organized general system of data management."2	2
Nelson's grand vision of a universal library and publishing system has come in for its share of derision. In 1995, the Wired magazine writer Gary Wolf devoted twenty thousand words to detailing what he called "The Curse of Xanadu." "Nelson's Xanadu project," he wrote, "was supposed to be the universal, democratic hypertext library. . . . Instead, it sucked Nelson and his intrepid band of true believers into what became the longest-running vaporware project in the history of computing—a 30-year saga. . . . [an] amazing epic tragedy. . . . [and] an actual symptom of madness." Nelson responded angrily to Wolf's profile, but he has also hinted that he views Xanadu as an impossible dream. He took the word from the imaginary home of Kubla Khan in Samuel Taylor Coleridge's uncompleted poem of the same name; Orson Welles (one of Nelson's heroes) used the same word for Citizen Kane's extravagant, uncompleted mansion.3	3
And yet, just five years after Wolf's obituary for Xanadu, the dream of a universal hypertextual library seems less like the narcotic imaginings of Samuel Taylor Coleridge or the fantasies of Ted Nelson than a description of a multibillion-dollar industry called the World Wide Web.4 Even those of us whose professional calling requires us to think soberly about the distant past need now to consider whether such a contemporary development will reshape the ways we research, teach, and write history. Can professional historians look forward to a future in which they can access all the documentary evidence of the past with the click of a mouse? How far have we already come toward reaching that dream?	. . .