In a 1945 paper for the popular media, Vannevar Bush proposed that science be put to use in organizing the vast record of human knowledge. Inspired by his previous work in microfilm mass storage, Bush envisioned an information workstation—the memex—capable of storing, navigating, and annotating an entire library’s worth of information. His idea of push-button linking between documents seems to be the forefather of modern hypertext. As described, Bush's memex was based on what was thought, at the time, to be advanced technology of the future: ultra high resolution microfilm reels, coupled to multiple screen viewers and cameras. The memex reflects a library of collective knowledge stored in a piece of machinery described in his essay as "a piece of furniture."
As We May Think predicted (to some extent) many kinds of technology invented after its publication, including hypertext, personal computers, the Internet, speech recognition, and online encyclopedias: "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." Bush envisioned the ability to retrieve several articles or pictures on one screen, with the possibility of writing comments that could be stored and recalled together. He believed people would create links between related articles, thus mapping the thought process and path of each user and saving it for others to experience. Wikipedia is one example of how this vision has been realized, allowing users to link words to other related topics, while browser user history maps the trails of the various possible paths of interaction.
Bush urges that scientists should turn to the massive task of creating more efficient accessibility to our fluctuating store of knowledge. For years inventions have extended people's physical powers rather than the powers of their mind. He argues that the instruments that are at hand which, if properly developed, will give society access to and command over the inherited knowledge of the ages. The perfection of these pacific instruments, he suggests, should be the first objective of our scientists.
So far, Bush’s acquisition and storage mechanisms seemed
reasonable, but they had a major shortcoming for retrieval of information:
images could not be searched and retrieved by machine. Bush recognized a need for digital indexing
of documents, that is, associating a machine-readable index code with each
document, so that a machine could identify and retrieve a document from mass
store automatically. He termed this
retrieval selection. Punch card selectors had existed for census gathering,
but Bush extended the idea to larger, faster systems. In the paper, he gives the example of a
department store accounting system, where each transaction combines punch cards
from the customer, salesperson, merchandise, and general accounts. The general accounts card(s) would have to be
selected and updated by machine in a back room, e.g. matching customer codes to update billing, e.g. matching merchandise codes to
update inventory.
I can envision a computer with artificial intelligence programmed so that it is possible to speak using vernacular of any given time in our history and determining the prognosis of illness with the current technology of electronic medical records in place. I foretell the role of the physician as information disseminator of prognoses determined by machine. And I can foresee a world in which our ethics and morality are foreshadowed by technology and perhaps artificial intelligence stepping in regardless of human desires, i.e., artificial intelligence trumping human intelligence.
