Are.na Influences is a series in which we discuss some of the ideas that continue to provide inspiration to us while building this platform together. In this post, Will Freudenheim revisits three twentieth-century thinkers who imagined computing technologies as a way to expand and explore the interconnectedness of human knowledge.


Every act of information consumption is as much about gaining knowledge as it is preparing for the loss of memory. This preparation implies rituals of storing and organizing information for later use, which for much of human history was done by writing in notebooks, cataloging in library indexes, or developing elaborate mnemonic devices. But the Internet provides us with an ever-widening landscape of available material, making the task of storing and organizing an increasingly difficult undertaking. Computer pioneers Vannevar Bush, Douglas Engelbart, and Ted Nelson anticipated this challenge, and each imagined computing technologies that would create new rituals for organizing and contextualizing information by embracing our capacity for association and to forge interwoven connections.

These projections of future research tools provided much of the foundation for what has become personal computing and the Internet. However, not all of the fundamental principles behind these technologies survived to today. Bush, Engelbart, and Nelson form a lineage of thinkers that viewed learning and research as a process of linkage, recontextualization, and rediscovery.

visions of connection

Vannevar Bush: The Memex in “As We May Think”

Vannevar Bush was an American engineer and inventor who led the U.S. Office of Scientific Research and Development throughout World War II. Overseeing the work of some six thousand scientists, the agency was responsible for the evolution of new medical treatments, analogue computing, advanced military vehicles, and, notably, the atomic weaponry research that became a precursor to the Manhattan Project. His work would both shift the face of global warfare and set the stage for the military industrial complex.

Like many scientists after the war, Bush shifted his focus towards technologies that could promote communication and understanding. Arguably his most influential post-war work took the form of an essay titled “As We May Think,” in which Bush speculated about future tools that could enable new modes of research and creativity. Through his work with the military, Bush reflected that while the pace of scientific research was rapidly developing, individual scientists were increasingly unable to discover and remember relevant information among the mass of inconsequential research. He began working through this problem by imagining a future desk assistant device which he called the Memex, a portmanteau of “memory” and “index.” Bush’s designs for the Memex combined a number of technologies that he expected would soon advance, including microfilm, viewing screens, and mechanized retrieval systems. The idea outlined in this paper for a desk-sized workstation device that could hold a flexible and browsable library of information became a model that has since informed and shaped the age of personal computing.

The Memex was a device not only for holding and displaying information, but one that allowed the user to create paths between documents that could be stored and retrieved. A Memex user would have two documents open at once: one for reading and annotating information, and one for looking through related documents that came to mind as the user researched. While exploring connections between documents, the user would use the Memex to permanently associate them, creating a link between the documents so that one item would immediately retrieve its linked partner upon reading. This mode of textual connection would later inspire the concept of hyperlinks.

Bush imagined users forming extended chains of these links, which he called “associative trails.” Informed by his understanding of human memory, these trails could break off and extend into side paths that could be as winding and elaborate as our own thoughts. Associative trails could also be passed among friends or colleagues to enable the distribution and extension of one’s associations. Bush believed that approaching research as a process of building common networks of interwoven associative links could increase one’s capacity for quickly discovering and grasping complicated concepts, and could offer an elevated starting point for developing solutions. As he wrote in his essay, “Wholly new forms of encyclopedias will appear, ready made with a mesh of associative trails running through them, ready to be dropped into the memex and there amplified.”

Douglas Engelbart: Concept Packets and Hypertext

Bush’s essay was a major influence for Douglas Engelbart, an engineer and inventor who worked from the 1950’s through the late 20th century in human-computer interaction. Though many of the tools Engelbart created became widely influential to the emergence of personal computing and the Internet, many of his ideas for how these tools should be used were less widely adopted. In his 1962 essay “Augmenting Human Intellect: A Conceptual Framework,” Engelbart outlines a grand vision for computing technologies where machines augment the human capacity for problem-solving on a societal scale. Like Bush, Engelbart believed that the first step toward this goal should be to address the limitations of an individual’s basic capacity for handling information.

visions of connection

One of Engelbart’s first experiments in augmenting human information handling was a simple analogue notecard system that he developed for his personal work. Inspired by Bush’s associative trail model, Engelbart’s “concept packets” used decks of IBM punch cards. Each card in a deck represented a fragment of an idea or concept, and would be notched with a number of holes along its edge, each hole representing a category. By threading a needle through multiple holes across the deck, one could retrieve all of the cards from the category or subcategory. Engelbart built detailed decks of these cards for research projects and manually linked and rearranged them to form conceptual frameworks. As he worked with this primitive system in the early 60’s, Engelbart imagined future machines that could automatically sort and visualize the meaning of linkages beyond the notches alone.

Engelbart looked to the limitations of physical concept packets as a starting point for developing advanced electronic systems that could accomplish his higher goals of boosting the “collective IQ.” Six years after writing “Augmenting Human Intellect,” Engelbart delivered what is now referred to as “The Mother of All Demos,” with a presentation of the oN-Line System that displayed the first prototypes of hypertext, collaborative word processing, the mouse, multiple windows, and videoconferencing. These tools remain Douglas Engelbart’s legacy. However, his abstract ideas about how these tools would enable people to work together more effectively and result in significant cognitive and social changes were never quite realized as he’d imagined them.

Ted Nelson: Hypertext and Xanadu

Ted Nelson was similarly concerned with the complexities of the creative impulse, and was also deeply influenced by Bush’s associative trails. Working independently of Engelbart, in 1965, Nelson described an idea for a flexible version of an electronic Memex research machine in his essay “A File Structure for the Complex, the Changing and the Intermediate.” His vision avoided hierarchical arrangements of files, instead enabling users to change and rearrange the contents of each file. The device would also use what he called “dynamic outlining,” where rearranging the outline of a text would automatically rearrange the main text as well. In the conclusion of this piece, Nelson coined the term ‘hypertext’ to describe a new system of nonlinear reading and linkage. He saw the future of hypertext as a medium that which would afford readers with different skills and interests to approach one ‘text’ in any variety of ways.

visions of connection

Ted Nelson continued to be driven by the fundamental belief that anyone engaging with material (graphic or textual) should be working with its “intrinsic structure.” What the thing is to the user should be the sandbox within which a person plays, rather than working within arbitrary set of conventions like a word processor. Nelson’s life’s work became working toward this goal with an immense project that he calls Xanadu, a system combining an incredibly flexible document editor and universal index. The utopic vision behind this system was that Xanadu could create an entirely new literary paradigm—one which would liberate thinkers from the constraints of static mediums and the challenges of publishing and copyright systems.

Inside of Xanadu’s ‘Docuverse,’ one would find a networked, living index which held the entirety of human writing. And within any given text in the Docuverse, the links embedded within would visibly and directly connect to their original source. As a text might point to a wide variety of quotations and sources, it would weave a tapestry of potential connections and contexts, each offering new worlds of possibility for research and exploration. This relied on a unique process of direct quotation—Nelson called it ‘transclusion’—which would lay at the core of Xanadu. Using transclusion, information could exist in multiple places at once, and as it was edited, both its updated and previous versions would remain connected to all other documents that reference it.

One’s documents would be eternally revisable in this paradigm, so that no one could ever have the last word. Rather, the documents and their linkages within Xanadu’s Docuverse would exist in a constant state of flux and conversation.

The numerous modes for interlinking information composes some of Nelson’s most fascinating work. One-directional ‘jump’ links—the kind that we are familiar with today on the Internet—are just one of a number of types of linkage that Nelson described. Documents could be viewed as parallel groups (as with the Memex), such that a user might draw connections between multiple side-by-side texts that would reappear as another reader scrolled through a text. Information could be threaded together as a series of “chunks” that could compose more complicated structures that were meant to be read nonlinearly. One could also build “compound documents,” where writing from multiple sources could be quoted in one context, each quote acting like a window into its original source which would sit a level below. The thinking of the users within Xanadu’s Docuverse was to be as liberated as the conventions of writing.

visions of connection

Nelson described his early plans for these ideas in his famous double-volume work “Computer Lib / Dream Machines” in 1974 as well as in his followup “Literary Machines” in 1981. Both pieces are now heralded as epics of the computer revolution. Of course, Xanadu did not develop into the world’s universal literary platform, but instead hovers like a ghostly shadow over the Web. Though his battle may be lost, Nelson still vehemently maintains that the Internet offers only a mere fraction of the functionality it should inherently make available to writers and thinkers. After working on the project for more than five decades, however, the prototypes of Xanadu remain under development.


As the world’s information continues to gush towards us, perhaps we should continually reassess our tools for wading through it. Bush, Engelbart, and Nelson’s ideas offer (what still feel like brand new) strategies for this kind of wading. Their shared belief in the incredible associative capacity of the mind and the power of sharing those associations continues to be influential. As we explore the intentions behind these projects, we continue to look toward developing a space inspired by what these thinkers imagined—one that enables users to create their own rituals for storing, organizing, and remembering.