Paul Otlet. He dreamed of an international network of electronic tools, documents and indexing more than seventy years ago. He dared dream of a network of information linked through symbolic code - at a time when most people could barely figure out a municipal transit schedule. He thought of the hyperlink fifty years before anyone could really make it work.

And then he convinced a government to fund his work. THAT is impressive. Do you know how hard it is to convince bureaucrats to give money to your visionary yet obsessive project? Especially if your project skews towards the crazy side of the innovative/crazy scale?
A brief history of Otlet and his Mundaneum can be found in the New York Times:

“… Although Otlet’s proto-Web relied on a patchwork of analog technologies like index cards and telegraph machines, it nonetheless anticipated the hyperlinked structure of today’s Web. “This was a Steampunk version of hypertext,” said Kevin Kelly, former editor of Wired, who is writing a book about the future of technology.

Otlet’s vision hinged on the idea of a networked machine that joined documents using symbolic links. While that notion may seem obvious today, in 1934 it marked a conceptual breakthrough. “The hyperlink is one of the most underappreciated inventions of the last century,” Mr. Kelly said. “It will go down with radio in the pantheon of great inventions.”

Mathew Ingram has some pointers as well.