Towards A Library Without Walls

A few weeks ago, fellow user Sam Hart posed the question in the Personals channel, “What might ‘librarianship’ mean for a platform like” Embedded within this question is the inverse—what might a platform like mean for librarianship?

It’s a question I consider often as a longtime user halfway through my MLIS degree, the accreditation commonly tied to working as a librarian in the United States. The field of library and information studies (LIS) circles around methods for the description, classification, organization, and access of information in all its forms. These issues are often taught through formalized and highly-structured knowledge organization systems, such as Library of Congress Classification and its attendant use of subject headings and call numbers, which are built upon a procession from the general to the particular and (implied) greater to lesser importance—all from the position of the U.S. as a place “from which the rest of the world is viewed.”1 Clearly, these systems are not without considerable frictions, not least of which include their emphasis on Western-centric, empirical, and scientific discourses, often at the expense of reflecting other systems of value and knowledge generation—be they non-Western, non-white, queer, indigenous, feminist, or otherwise.

As Martha Lampland and Susan Leigh Star observe in their essay “Reckoning with Standards,” “one person’s infrastructure is another’s brick wall.”2 This is apparent when considering rigid, hierarchical, and text-based systems like LCC in the context of arts librarianship—my own personal interest—which could most benefit from discovery tools that can accommodate abstraction, visual resources, and serendipitous browsing. Rachel Clarke reports in her essay “Cataloguing for Art and Design School Libraries” that “compared to traditional library users, artists favour ‘peculiar search techniques,’” and library services should be informed by a meaningful understanding of their information-seeking behavior.3 A hypertextual platform like then can show us what modes of knowledge production and sharing elude the tools already in place in libraries, offering alternative strategies for organizing and sharing information.

In my experience, it’s not uncommon to encounter variations on the hypertext in academic coursework designed for librarians and archivists, including systems proposed by Vannevar Bush (the Memex), Paul Otlet (the Mundaneum), and Ted Nelson (Xanadu). In his article “Visions of Xanadu: Paul Otlet (1868–1944) and Hypertext,” W. B. Rayward defines hypertexts as consisting of two elements—nodes and links—that can be manipulated by the user and serve as the elements of a system in lieu a continuous, linear flow of textual content.4 (In the case of, blocks serve as nodes and links operate through channels.) Today, the Memex, the Mundaneum, and Xanadu are often referenced as precursors to the present-day Internet, and all three are regularly surface in channels alongside other examples of associative strategies such as Apple’s HypercardWendy Hall’s Microcosm, the Japanese practice of Zuihitsu, and the poetry of Stéphane Mallarmé.

Given the many precedents of hypertext throughout the history of knowledge organization, the question of’s relevance to librarianship is not whether associative indexing has contemporary relevance, but rather what qualities does the knowledge generated in this hypertextual context possess as a result of this approach? And what potential does a hypertextual tool have to serve alternative forms of knowledge?

Many of the potentials of a tool like can be drawn out in its similarities and contrasts to Vannevar Bush’s formative concept of the Memex, which Bush conceived in his 1945 Atlantic Monthlyarticle, “As We May Think.”5 In a previous post on this blog titled “Visions of Connection,” Will Freudenheim explores the Memex in depth, describing it as a “desk-sized workstation device that could hold a flexible and browsable library of information,” a model which has “since informed and shaped the age of personal computing.” As Internet users, we encounter Memex-like tools on a daily basis, even in something as simple as our browser history, which tracks browsing in a “trail”-like, automatically-generated list of URLs punctuated by timestamps. Based on the many tools and experiences that inherit the intellectual lineage of the Memex and other forms of hypertexts, we can consider the implications of associative thinking on information and the generation of knowledge. Of particular interest is a closer reassessment of the sustained emphasis on the forms of scientific, singular, and ultimately universalizing conceptions of thinking that dominate the intellectual touchstones of LIS at the expense of more expansive visions of knowledge production.

In “As We May Think,” Bush’s argument comes in the wake of two World Wars and is largely concerned with science as the cornerstone of knowledge as it relates to power and elevating the spirit of man. This kind of universalism ignores the aforementioned alternate value systems, as well as other significant forms of knowledge (i.e. ethical, theological, philosophical, cultural, musical, or artistic), and historical narratives that might call into question his progressive vision of modernity. By limiting his conception of associative indexing to the production of scientific knowledge, Bush misses the potential that these other forms of non-empirical knowledge—where definitive truth is elusive or potentially even undesirable—could find even greater benefit from hypertextual models. With, the content in many of its user-generated channels evades the noble purposes Bush envisioned for his Memex. Instead, the mundanethe mimeticthe frivolousthe artisticthe irreverentthe visual, and the ephemeral come to the fore. Here we can recognize associative coalescing as a form of non-result-oriented knowledge production, where the fuzzy boundaries of a subject are collectively defined through free association and family resemblances as connections between channels and blocks are improvised over time.

Collaboration has also become key to the way we conceive associative indexing on today’s version of the Internet, which could not have been anticipated by Bush at today’s scale. In “As We May Think,” Bush does acknowledge the possibility of sharing links generated by the Memex in the example of a researcher reproducing a trail on the Turkish bow for inclusion in a colleague’s “more general” trail.6 However, the scale of a hypertextual tool such as, which has over 20,000 users, far exceeds the one-to-one exchange Bush envisioned for his Memex, with significant implications for associative indexing. This phenomenon has its own neologism, “crowdsourcing,” wherein large numbers of users, most typically through the Internet, contribute to an information platform, as seen widely from commercial endeavors such as Google-owned Waze to non-profit projects such as Wikipedia. The relative advantages and disadvantages of crowdsourcing for knowledge production are the subject of much literature but could be briefly alluded to here in terms of diversity of material, collective intelligence, increased scale, and lack of consolidated control. But at its most promising, crowdsourcing creates the potential for rich communities that can form around information sharing, as is well articulated by Paul Duguid and John Seely Brown writing on the social life of information:

“[D]ocuments do not merely carry information, they help make it, structure it, and validate it. More intriguing, perhaps, documents also help structure society, enabling social groups to form, develop, and maintain a sense of shared identity. Viewing documents as mere information carriers overlooks this social role.”7

Considering the ways in which operates within a community of artists and culturally-engaged individuals, contrasting with Bush’s Memex highlights the importance of conceiving how knowledge forms, knowledge tools, and knowledge communities all interplay with one another. By acknowledging other forms of knowledge beyond the scientific and better understanding the role sociality plays in our contemporary experience of information, we can better define what constitutes information and how best to describe, classify, organize, and make it accessible as librarians. Rather than prioritizing static information, fixed organization, and solitary experiences as the conventional library environment is known to do, those of us who work in LIS can adopt the more boundless strategies that we encounter in hypertextual tools such as for the benefit of the communities that we serve, essentially working towards becoming a library without the brick walls that Lampland and Star refer to in regards to infrastructure that fails to serve user needs. Parallel to thinking about what might mean for librarianship, we can look to extant projects such as the Prelinger Library and the Sitterwerk’s Kunstbibliothek, whose methods for organizing their material also exist as an alternative to more traditionally-organized libraries.

So to expand on Sam’s question and its inverse: What could a reference interview that uses look like? What would happen if books in an OPAC were nodes that could be linked by users? And what if the discovery tools we design actually encouraged research that is social, elusive, and nonlinear?

Karly Wildenhaus is a researcher living in Los Angeles.