Bush, Licklider, and Nelson

From Gerald R. Lucas

The pieces by Bush, Licklider, and Nelson in the New Media Reader (see Foundational and Transitional Thinking) seem to continue expounding upon the question that Turing and Wiener were interested in: just how do humans think and what does that mean for the design and use of technology? Just what is a well-designed tool that will allow scientists and, by implication, the rest of us work in the most productive way that we can. Are there generalizations that can be drawn about how humans work, or is the only constant involved in human thought the notion of change? Bush, Licklider, and Nelson are interested not only in how the individual records her knowledge, but how then she shares it with the rest of the world. Computing technology, suggest all three, might hold the answer.

Bush, in his essay “As We May Think,” invents what he calls a “memex,” or a sort of Victorian contraption with wheels, pulleys, and switches that can catalog a person’s interests in such a way as to provide a record, or “trail,” of knowledge through association. Bush sees the human mind working not necessarily by reason, or though predictable algorithms, but by association and chance that forms a “web of trails carried by the cells of the brain.”[1] His “memex” is “a device in which an individual stores all his books, records, and communications [can’t you hear the RIAA and MPAA groan?], and which is mechanized so that it may be consulted with exceeding speed and flexibility.”[2] Yet, unlike the finite capacity of human memory, the memex can enlarge and make permanent the transitory pathways of neurons in the brain.

Bush’s Memex

Bush’s memex, as he conceives of it, is an analog machine, replete with little cameras that take detailed snapshots of materials on microfilm. However, his quaint vision right out of Babbage’s century need not be a representation of crude writing implements and dead media, but may bypass the senses all together, so that no translation would be necessary: since all forms from the external world are transformed to various electrical currents, might it not one day be possible to connect these currents directly to the currents that make up the human nervous system? Bush seems to think so. His memex, then, becomes a literal memory augmentation, so that a human’s “excursions may be more enjoyable if he can reacquire the privilege of forgetting the manifold things he does not need to have immediately at hand, with some assurance that he can find them again if they prove important.”[3]

Another fascinating aspect of Bush’s essay was his passing suggestion that human languages are not particularly adapted to communicate with machines. Instead of suggesting ways to design the machines to deal with this, he suggests that language will inevitably change. Here begins a disturbing notion of the humans changing for the technology, and not vice versa. I’m reminded of the marketing for some Intel gadget which said “You’ll find a reason to use it.” Well, why should we? This seems to suggest that how we want to work takes a secondary consideration to how the designers (and capitalists) may want us to work. Perhaps if computers have difficulty with human languages, they need to be revised, not us. I’m sorry, I do not want to have to learn a new way of writing just to communicate with my PalmPilot — I can’t even type, for goodness sake.

Licklider’s “Man-Computer Symbiosis” suggests that we should live in a symbiotic relationship with our computers, for we will each benefit the other, but does not address necessarily how we can do this. Computer keyboards and monitors are just not enough for a symbiotic relationship. I kept thinking of Star Trek’s computer working with Geordi to solve a crisis. The computer was able to create a likeness of another human — one that Geordi was physically attracted to — to work with him on solving a particularly difficult problem, one no doubt that would have caused the destruction of the galaxy if they were unsuccessful. This type of relationship seems to be what Licklider is going for: one that uses the strengths of each to augment the other’s weaknesses. Good idea. How do we do it? We’re back to interface and design. Maybe Siri is that initial step? Again, it seems that the closer we can design a computer to act human, the more symbiotic our relationship will be to it.

Ted Nelson, Feb. 19, 2011.

Nelson’s “A File Structure for the Complex, the Changing, and the Intermediate” tries to envision an interface, or at least a structure, that would allow humans to organize their digital lives in any way they wanted. Rather than a hierarchal structure, Nelson’s ELF (evolutionary list file) would be organized according to the user’s desires, and it would be able to evolve as the user does. I like Nelson’s idea, but his essay becomes bogged down in the abstract; I don’t think he even had a good idea of just how something like an ELF could be implemented.

What’s most interesting about Nelson’s piece is its conclusion: he posits his “philosophy” that complex and flexible file structures will enable the creation of new media, “the hypertext and hyperfilm,” and will facilitate the ever-evolving order of the human.[4] What Nelson means by “hypertext” is not its Web manifestation of today. Our hypertext is not really hyper because it can be organized linearly. Nelson seems to desire something more sophisticated — something that works symbiotically with the user to create a truly interactive and evolving form: “it has become possible to create a new, readable medium, for education and enjoyment, that will let the reader find his level, suit his taste, and find the parts that take on special meaning for him, as instruction or entertainment.”[5] Indeed, no human writer (programmer) could construct all of these varying contingencies; therefore, the only conceivable way to write these environments is to allow the computer a role in the creative process. Not only would this happen for artistic expressions, but also for the mundane structures of our lives. The medium must be flexible enough to go anywhere we want; Nelson continues:

Last week’s categories, perhaps last night’s field, may be gone today. To the extent that information retrieval is concerned with seeking true or ideal or permanent codes and categories — and even the most sophisticated “role indicator” syntaxes are a form of this endeavor — to this extent, information retrieval seems to me to me fundamentally mistaken. The categories are chimerical (or temporal) and our categorization systems must evolve as they do.[6]

Indeed, what is literature or art or philosophy but “categorization systems”? These systems change over time, and in this postmodern zeitgeist, they are not necessarily trying to find the truth. No one view is any more correct than any other; wouldn’t it be great if our computers mirrored this idea?

It seems, however, that the trend is going the other way: you must conform to whatever the current popular trend is, or you will be invalid. I think Nelson would be appalled at the notion of a thinking monopoly. Indeed, how does Windoze, iOS, Google influence and structure how we work, how we think, who we are? Instead of the computer dictating who I am, I would prefer to be the one dictating.


  1. Bush, Vannevar (2003) [July 1945]. "As We May Think". In Wardrip-Fruin, Noah; Montfort, Nick. The New Media Reader. Cambridge: The MIT Press. p. 45.
  2. Bush 2003, p. 45.
  3. Bush 2003, p. 47.
  4. Nelson, Ted (2003) [1965]. "A File Structure for the Complex, the Changing, and the Intermediate". In Wardrip-Fruin, Noah; Montfort, Nick. The New Media Reader. Cambridge: The MIT Press. p. 143.
  5. Nelson 2003, p. 144.
  6. Nelson 2003, p. 145.