Published in 1945, but originally written in 1939, Vannevar Bush’s article, “As We May Think”, is a fantastical description of an imaginary information machine called the “Memex” (Buckland, 1992) (Bush, 1945). Bush argued that existing methods of transmitting and reviewing the results of research were totally inadequate and that if a record was to be useful to science is must be continuously extended, stored, and above all consulted. The Memex was designed as a self contained desk, storing information through photographic technology, displayed on screens and also by utilizing associative trails to connect concepts, similar to how the brain works (Bush, 1945). So was Bush’s vision really such a thing of fantasy? His intent was that the Memex should operate “as we may think” as opposed to how other conventional inadequate information indexes worked. The machine itself, never caught on, but the concepts behind it offer a unique perspective on the development of information storage and retrieval systems.
Today, the dominant way in which information is retrieved is via the world wide web. How do today’s methods, then, compare with Bush’s vision for the future, represented in his Memex machine? The first distinction to be drawn in an examination of Bush’s vision, was the aforementioned reliance on microfilm which, at that time, was necessitated by a lack of any technology for digital mass storage (Buckland, 1992). The reliance on microfilm as a storage medium meant that there was no possibility for networked information between Memex machines. This was likely the greatest limitation of the machine, and an interesting point to consider when contemplating Bush’s vision of the future as being profoundly progressive. It marks an interesting dichotomy between his forward thinking and inability to look beyond conventional storage methods. By having each Memex user restricted to the information contained within their own machine, a state of isolation is created which does not foster an environment where information can be easily disseminated and build upon. Indeed, Bush’s reliance on experts as sources of Memex information, as opposed to a network of users themselves also represents a manifested contrast resulting from this isolation. It can be argued that one of the strengths of information found online today is that it comes as a result of dissenting opinions and contrasting perspectives. The democratization of information has made the culmination of everyone’s experience and knowledge vastly more important than that of a single ‘expert’. The flip side, however, is that the requirement of readers to be critically receptive of what they are reading has become more vital than ever. Because anyone can author a webpage, the currency and validity of that source should be viewed with the utmost scrutiny.
One of the defining features of the Memex was what Bush described as ‘trails’ – a newly conceived method of connecting different streams of thought or topics together which could be recorded and recalled at a later date (Bush, 1945). However, again, in the absence of networking there lies a notable limitation in the utility of trails. Memex libraries were replicated and stored locally for each user, and trails were constructed independently, to be shared only by hand. The ability to share trails through a network would have greatly expanded it usefulness. In addition to only having locally saved trails, Memex machines were further limited by their inability to refine searching using multiple search parameters and filters. This is a defining feature of web browsing today.
Trails, however, were not without their own merit. Because each user could custom tailer their own trails, this activity could be innately personal. It has been proposed that the concept of the trail was the forerunner of the modern hyperlink. The personal nature of the trail, however, would appear to be a feature which is not matched by the hyperlink. Hyperlinks are established by those who create webpages and are typically un-editable by those who read the pages. This utility can be worked in other ways though, such as through bookmarking.
One of the main arguments Bush had in forwarding his idea of a Memex machine was that at the time of his writing there was a proliferation in the availability of easily manufactured, reproducible and interchangeable parts, thus making the assembly and cost of machines routinely easier (Bush, 1945). The historical cost and inability to repair easily machines of the past was posited as one of the main reasons that collection and organization of information in a device such as them Memex has previously been impossible. While we do, indeed, have now a comparably inexpensive capacity to manufacture identical parts for computerized machinery, there exists now also what is now termed a ‘digital divide’ between cultures which have easy access to affordable technology and those which do not. This would be a factor Bush would not have had to consider in his proposal. So while the cost has certainly dropped, and the ability to manufacture on a grand scale has grown, the available to all areas of the globe and all class of people is not uniform.
There is no doubt that Bush was a visionary. His Memex machine did, indeed, represent a prelude to the invention of information networks, complete with digitized trails now known as hyperlinks. Bush’s vision, however, lacked several key innovations present in today’s information workstations, including searchable, digital content and rapid information sharing on a network. As highlighted by Caspi et al (2004), Bush “discounted the promise of digital information and of rapid information sharing by network. As such, Bush fell victim to a common flaw of technologists, to overestimate the impact of a technology in the near term and greatly underestimate its impact in the long term” (Caspi et al, 2004). Hindsight is always 20/20. Bush’s vision of a future where unprecedented information storage and retrieval was possible, while flawed, is still examined today because it was such a unique and profound perspective for his time.
Bolter, Jay David. (2001). Writing Space: Computers, Hypertext, and the remediation of print. Mahwah, New Jersey: Lawrence Erlbaum Associates
Buckland, M. K. (1992), Emanuel Goldberg, electronic document retrieval, and Vannevar Bush’s Memex. Journal of the American Society for Information Science, 43: 284–294.
Bush, V. (1945). As We May Think. The Atlantic Monthly, 176(1), 101-108. Retrieved November 8, 2011, from http://www.theatlantic.com/doc/194507/bush
Caspi, E., Shankar A.J., & Wang, J. As We May Think, Vannevar Bush. Course notes for CS294-6: Reading The Classics by Christos Papadimitriou, University of California, Berkeley, Fall 2004.