The Little Database

When asked for an example of the types of media-reflexive transformation I’m most interested in studying, I first turn to the PennSound MP3 file of William Carlos Williams’s “The Defective Record,” a poem that demands the concept of the gramophone even from the page.10 The final lines of the poem are printed: “Level it down / for him to build a house // on to build a / house on to build a house on / to build a house / on to build a house on to . . .” First recorded by Williams in 1942, these concluding lines call the skipping record into conversation with the destructive forces of mechanical reproduction. In his reading of the poem, Williams ignores the line breaks of the printed version and instead performs the metaphor of a defective record by mimicking his own vocal patterns in the repeated phrase, “to build a house on.” Simulating the all-too-common scratched repetitions in analog records, Williams baits the listener into realizing the track itself is not, in fact, “defective,” but rather presenting a material metaphor on the mechanisms of industrialization. Playfully amplifying the potential for a mechanical failure of the groove, Williams tricks his listeners in order to solicit a moment of productive confusion. This moment, this play with the meaning-making potentials of the material recording itself, is a hallmark feature of media-reflexive arts in the tradition of modernism and the historical avant-gardes. These playful encounters with media, I argue, sound out the otherwise inaudible effects of digitization.

For example, once “The Defective Record” is transcoded into an MP3 file, the poem’s own reflexive gesture changes to accent instead the work’s digital formatting. Audio files on the internet, of course, don’t regularly “skip” in the same way that a vinyl record might. The repetition of the line “to build a house on” finds a new foundation built on the MP3’s compression algorithm, dramatically leveling the wave-like stream of the analog wax track and contradicting the originally plotted effects of the poem, with regard to the play of both politics and aesthetics. The material metaphor of the scratched record is itself rendered defective in the face of modular digital playback. If there are “no ideas but in things,” as Williams first declared in a 1943 issue of The Old Line magazine, then what happens, we might ask, to the ideas when the things themselves rematerialize? Unlike the anticipatory readings performed by media archaeological studies that return to latent digital potentials found within historical works, The Little Database argues for ways of understanding the poem as a functioning new media object, operating within the parameters of its digital format and new media context.11

Beyond this point, there is much more at play in this file than a narrative of the formal and prosodic properties of the work, even if we track these changes from the page to the vinyl record to the MP3 and back. Reeling away from a close reading, to properly understand “The Defective Record” online, we must simultaneously consider the file playing alongside thousands of corresponding digital objects on sites like PennSound and UbuWeb, or indeed streaming to earbuds connected to my mobile device, or nested within the haphazard file folders that organize my laptop’s hard drive. Inhabiting the vexed space between preservation and distribution, between memory and practice, little databases reconfigure the contemporary experience of historical artifacts, and thus transfigure interpretative methods for each work in turn. More than simply degrade or distort analog works of avant-garde art and literature, digital forms of archival collection and circulation dramatically rematerialize historical works for the use and reuse of untold numbers of contemporary users. The same could be said for how our mixed-media machines remediate analog culture at large. In this way, the failures, glitches, errors, slippages, and unlikely reversals of media-reflexive works enable the reader to glimpse, for a moment, the otherwise invisible structures that seek to conceal these transformations within the increasingly high-fidelity formats of cultural transcoding.

How then might we attune critical frequencies to channel the significance of “The Defective Record” playing today? I argue that a close reading of transcoded objects can occur only in cooperation with playful performances of interpretation that weave together relational databases, variable formats, and the contingent array of technical and cultural mechanisms that facilitate these files. In considering how these systems operate, I draw from recent scholarship on critical code studies, comparative media analysis, and data justice. Every display of a digital object is a unique performance.12 Streaming bits are inscribed and reinscribed in hardware.13 “Invisible” code is parsed into visual or aural display.14 Interfaces and protocols control the expression of these streams.15 Users and editors rip, post, and repost files in a variety of online contexts.16 Updates render old formats inoperable.17 Bias lurks in algorithms, data sets, and infrastructures alike.18 Digital repertoires of care maintain, collect, and inflect each work in turn.19 And so on it goes, in the continual flux of technological development and the ongoing versioning processes of the always-on, always-developing, internet. Even a scattered list of variables exhausts the interpretive equation. Each of these emergent properties plays into any reading of the instantiation of a digital object. I argue that such a scenario demands scholarship that is bound to exploratory methods of interpretive performance and contingent reading. The Little Database navigates these contingencies of meaning-making by exploring an array of digital formats and the features of online collections, occasionally to the pointed exclusion of reading the “content” of specific files. Scaling through methodological frameworks built to encapsulate interpretation at different orders of magnitude enables a poetic reading practice capable of traversing urgent concerns not accessible in seclusion.

In this way, it has proven useful to analogically cast online collections as “little databases” after the little magazines that served as a vehicle of modernism and the historical avant-gardes.20 Periodical studies on the little magazine offer an array of methodological perspectives developed to address precisely the challenges of writing about heterogeneous websites that incrementally release new materials over time. That is to say, no magazine is defined by any single work included, but every work is inflected by its place within the social and material texts of the magazine. In addition, the sites I examine update the defining characteristics of the little magazines. Eclipse, for example, is driven by a set of arguments constructed by its founding editor, Craig Dworkin, whose archival interventions engage in contemporary aesthetic and literary practices through a provisional platform that develops in periodical increments over time.21 Like the study of the full run of a magazine, these databases require a confluence of close, hyper, and machine reading practices.22 These same reading practices could be extended not just to a range of public little databases online, but also to the private collections that inhabit user hard drives everywhere.23

The relation of the little magazine to the contemporary database is also indebted to trends in periodical studies that Robert Scholes and Cliff Wulfman have described as a transition “from genre to database.”24 In all these regards, the field of periodical studies presents an array of methodological and critical interventions that helpfully map onto the chaotic landscape of digital collections and online repositories. Despite the large number of objects hosted on these sites, it should also be noted that, from a computational standpoint, the data they contain, such as hundreds of issues of a little magazine, remain eminently computable. The question is why and how computational tactics might reveal, demystify, or disorient these collections. In each chapter, I explore the promises, pitfalls, and potentials of creative modes of computation—not only to better grasp these sites, but to extend an understanding into the personal collections and ordinary media practices that structure the experience of everyday network culture.

Resisting analytics, I contend that the little database instead courts varieties of creative research that might facilitate a poetics of computation, engaging with each idiosyncratic collection as a site-specific model for meaning-making activities. Following the logic of the little magazine for these experiments, I examine how these little databases have forged their own “periodical codes,” inventing time-based tactics for circulating digital objects as patterns of use and standards of access continue to develop online.25 When reading a little magazine, a combination of macro- and microscopic modes of analysis yields valuable insights not readily available to either method in isolation. Moreover, like pieces featured in a magazine, the transcoded work can be read only through its instantiation within the social and technical networks that deliver it, in a given moment of delivery. By the same token, in a kind of hermeneutic circle, to read the database we must enter through the individual works that are the collection’s contents. In both directions, social networks, technical protocols, and cultures of use inscribe the limits of potential readings. In the second chapter, I pay particular attention to how the historical challenges faced by periodical studies might lend insight into the manifold difficulties of accounting for the little database on the internet.

On the other side of “little,” this study of transcoded objects cannot proceed without acknowledging Manovich’s now-historical account of the database as a cultural form in The Language of New Media. A database can be simply defined as any organized collection of data. From relational modeling in SQL (structured query language) to mundane tabulation in everyday spreadsheets, a wide variety of database models structure how data is organized, stored, indexed, displayed, and manipulated. When it comes to the indexical accumulation that undergirds the internet, Manovich argues that “every site is a type of database.”26 For technical purists, this statement might be something of an exaggeration, just as it’s clear that the sites I examine extend far beyond the little magazine in accordance with its stricter definitions.27 The play between these definitional boundaries serves as a guide to the grey area of the online collections I examine and how they relate back to ordinary hard drives everywhere. Online collections hover between periodical publications and structured databases. The tensions between site and database are most fully explored in the fourth chapter, which considers the database logic inherent even to the construction of a single digital compilation movie.

From the outset, however, my introduction of a “little” difference aims to disrupt some of the claims Manovich advanced in 2001. In particular, I depart from his assertions concerning the antagonism between database and narrative, the linguistic parallels between paradigm and syntagm, and the reading of the database as the dominant “symbolic form” through which contemporary culture reproduces itself.28 Following the critique that N. Katherine Hayles levies in her book How We Think—to pick one among the many, which chart back to a forum at PMLA in 2007—my notion of the little database clusters instead around an unstable symbiosis between the narrative (and poetic) structures of the little magazine and the relational juxtapositions of the object-oriented database.29 To distort a biological metaphor deployed by Hayles in this work, I explore the “technogenesis” inherent to the evolving relations between the life of an artwork and its digital afterlife. Further, I argue, the significance of this continuously evolving feedback loop can be accessed and addressed only at site-specific moments in time.

This practice of tracking the material instantiations alongside transcoded versions of works within the little database owes much to the bibliographic specificity of comparative media analysis developed by Hayles.30 Following the bibliographic attention that media scholars like Lisa Gitelman, Johanna Drucker, Lori Emerson, and others pay to the history of the book within digital media contexts, I argue for a “material text” reading of these formats.31 In this regard, the variable processes of online collections might be recoded in terms of circulating versions of format-based editions. Appending their digital versions to a long series of historical iterations of these works, I argue for a renewed attention to the technical specificity of formats and the contextual effects of little databases in use. Here, once more, the media-reflexivity of the works in the little database I study enables methods of what Hayles terms “media-specific analysis” that attend “both to the specificity of the form . . . and to citations and limitations of one medium in another. Attuned not so much to similarity and difference as to simulation and instantiation.”32

This comparative practice requires a concept of interpretive change over time. In the title essay to How We Think, Hayles further develops a “theoretical framework in which objects are seen not as static entities that, once created, remain the same throughout time but rather are understood as constantly changing assemblages in which inequalities and inefficiencies in their operations drive them toward breakdown, disruption, innovation, and change.”33 Building on this frame, the little databases I examine have variously ceased operations (Textz), faced repeated dramatic interruptions (Eclipse), suffered the collapse of functionality (Mutant Sounds), supported a file format that has since been cancelled (UbuWeb), or faced innumerable breakdowns and disruptions (EPC). In these lapses, these glitches, the smoothly operating protocols that undergird little databases may be brought to comparative analysis and bibliographic scrutiny.34 Just as “The Defective Record” reveals itself when it skips, the collection comes into view upon its breakdown. The Little Database lingers with these glitches to discover moments in which these “constantly changing assemblages” caught between old and new media formats might reveal themselves in the fissures.

Contingent Reading

The changing assemblages that constitute the little database hinge on a notion of contingency. Each chapter of The Little Database passes a series of “anecdotal” interpretive filters across the respective little database it queries; a tactical approach to what Alan Liu has termed, in plural, “contingent methods.” Best encapsulated by Liu in the introduction to his collection of essays Local Transcendence: Essays on Postmodern Historicism and the Database, this contingency is rooted in the database format itself.35 Liu traces the processes by which pervasive information management practices proceed to disengage form from content through database queries, which care little for either in a traditional sense. Contingency in this context demands a shifting tactical method, understood as the “methodical tangency of postmodern historicism” rooted in feedback modulations between the past, the present, and the future.36 This tangency is marked by a “zigzag mode” of interpretive engagement wherein the posthermeneutical characteristics of contingency facilitate scholarship that unfolds across unanticipated, serpentine directions, in relation to the mediations at hand. In both the academic prose of these chapters and the expanded poetics described in the interludes, I follow Liu’s recommendation to perform “the act of such mediation through actual media innovation or allusions to such innovations in its own form, thereby methodically bringing to view a sense of simultaneous sameness and otherness in our relation to history.”37 The contingency of the digital object is mirrored by the variability of methods and formats in the chapters and creative works described in this study.38

Just as the database introduces contingent relations to history, the cultural practices of the internet introduce transformations to how we might understand historical works subject to variable formats of digitized circulation. Hito Steyerl sketches the turbulence inherent to the variable lives of compressed files in the foundational essay “In Defense of the Poor Image.”39 For Steyerl, poor images exemplify the freedom of gift economies, on the one hand, while finding themselves caught up in the “vicious cycles of audiovisual capitalism,” on the other.40 The article summarizes popular discourses concerning the circulation of transcoded digital objects, offering an excellent précis of current discussions on the highly compressed formats that proliferate in the collections I examine:

On the one hand, [the poor image] operates against the fetish value of high resolution. On the other hand, this is precisely why it also ends up being perfectly integrated into an information capitalism thriving on compressed attention spans, on impression rather than immersion, on intensity rather than contemplation, on previews rather than screenings. . . . It transforms quality into accessibility, exhibition value into cult value, films into clips, contemplation into distraction.41

Focusing on degraded forms of distribution, copyright politics, and pervasive qualities of cultural distraction, Steyerl presents a core paradox of file-sharing networks: the poor image is about “defiance and appropriation just as it is about conformism and exploitation.”42 By this account, the compressed file is a debased rendering of the original work, a “ghost of an image,” interesting only in terms of its politicized accessibility or cultural indexing. Certainly, these forces are at play in the files I examine. However, alongside variables like speed and distraction, simulation and defacement, I argue for a radical revaluation of the meaning-making potentials inherent to compressed files. These “poor images,” I contend, are significant aesthetic objects in their own right, with complicated (and fascinating!) stories well worth unraveling beyond concepts of copyright politics and degradation. A poorly captured JPG can inspire hours of longing just as a badly ripped MP3 of a poetry reading can warrant endless repetition.

The temporality of attention we might pay to any “poor” digital object might best be encapsulated by renewed critical attention to an economy of care in recent media scholarship. Increasingly, all our most meaningful encounters are mediated by the formats that populate our networked devices. Of course, we might also discover more affectively pressing forms of meaning inhering within the affordances of these same media formats. In league with Nanna Bonde Thylstrup’s conclusion to The Politics of Mass Digitization, among others, I agree that we must move past “the predominant tropes of scale, access, and acceleration in favor of an infrapolitics of care—a politics that offers opportunities for mindful, slow, and focused encounters.”43 Especially now, as the era of always-on streaming media produced by mass digitization works to obscure the infrastructural formats of the digital objects we encounter, attending to the infrapolitics that shape those cultural encounters demands renewed attention and redoubled care. In other words, how might we see what we’re seeing when we experience transcoded works online? At what points do the material specificities of these meaningful aesthetic experiences come into view?

Just as the operating systems and social networks of media platforms are subject to a continual state of development, the relatively stable conditions of protocols and standards present sturdy entry points for revealing layers of media formats in these experiences. Recent works on standards, infrastructures, and formats inform my approach to the localized transformations of digital objects.44 Most directly, this project owes much to the “format theory” developed in Jonathan Sterne’s MP3: The Meaning of a Format. Sterne urges media scholars to consider the cultural, social, and historical development of standard formats as a way to “modulate the scale of our analysis of media somewhat differently. . . . Studying formats highlights smaller registers like software, operating standards, and codes, as well as larger registers like infrastructures, international corporate consortia, and whole technical systems.”45 By attending to standard formats for text, image, sound, and movies on the internet, this study of digital objects works to ground the indeterminacy of medial formations and variable conditions of display. Sterne emphasizes the political urgency of this position in his conclusion: “Media remain on the scene, but they are diluted. . . . Future confrontations over democratic media systems and the right to communicate will be held over infrastructures, protocols, formats, portals, and platforms.”46 As Tung-Hui Hu, Safiya Umoja Noble, and Wendy Hui Kyong Chun have variously demonstrated, the political stakes of revealing the workings of digital infrastructures, algorithms, and formats have never been more urgent.47 In each approach, an engagement with the political reality of networks requires modulating between what appears on the screen and the expansive, often invisible sociotechnical systems that code these appearances.

Working this network description down to the finer details of specificity, I turn to my coeditor at Eclipse, Craig Dworkin, whose practice of “radical formalism” has proven essential to each contingent engagement. Dworkin defines radical formalism in Reading the Illegible and applies the same reading tactics to Eclipse in his article “Hypermnesia,” which I explore in chapter 2. Borrowing from Leon Roudiez’s “paragrammatics” and Ihab Hassan’s “misreading,” Dworkin pursues “the closest of close readings in service of political questions,” considering “what is signified by its form, enacted by its structures, implicit in its philosophy of language, how it positions its reader, and a range of questions relating to the poem as a material object—how it was produced, distributed, exchanged,” wherein “form must always necessarily signify but any particular signification is historically contingent.”48 This radical formalism informs Dworkin’s appraisal of the Eclipse archive, which is in turn augmented by a range of conceptual models for the dynamic procedures and material processes of digital objects. By way of its interludes, The Little Database seeks a radical formalism applicable to historical artifacts appearing as new media objects, within a set of historically contingent and contextually networked databases. The performance of interpretation is reductively sketched within the highly lossy compression format of an introduction: poetics find meaning in practice. In other words, this methodological blueprint relies on specific enactments of a media poetics.

A Poetics of Media Formats

Even while I focus on forces of dispersion in these pages, I aim to perform a great deal of synthetic activity. If the database as a form encourages contingent methods of scholarship, the samples from the scholarly database itself require variable, tactical uses geared toward specific medial situations. This book’s methodological framework is indebted to a wide array of thinkers across fields including material text studies, the digital humanities, comparative media analysis, affect theory, critical code study, literary and art history, poetry and poetics, cinema studies, and cultural theory, among others.49 As the title of the book indicates, my trajectory through this thicket can be provisionally categorized as “a” poetics of media formats. Following the foundational arguments laid out by Adalaide Morris in New Media Poetics, I consider this media poetics “an ongoing, elastic, and capacious process rather than a taxonomically precise product: as befits the processual or process-driven nature of computers, the emphasis . . . is on the act of making rather than the thing made, on forces rather than stable formations.”50 Locating each process of making as a singular path of discovery within an unstable field, the arguments of The Little Database explore the meaning-making activities of media formats and the possibilities for poetic thinking in media contexts new and old alike.

Indeed, I believe that a contemporary poetics practice can be articulated only through a rigorous consideration of coterminous writing technologies in all their complicating processes of mediation and remediation. As they say: “Our writing tools are also working on our thoughts.”51 This poses significant challenges to the intensive study of even the most mundane objects available to the contemporary poetics scholar. Not only is media built on the technoscience of programmers, mathematicians, and engineers, but the fields that plug into the cultural, political, and aesthetic aspects of media formats are so enmeshed in specialist discourses and disciplinary traditions as to muddle any comprehensive method. Given these conditional demands, my study necessarily surveys a widely interdisciplinary array of scholarly output in order to address the little database in variable configurations.

While methodologically indebted to the constellation of texts outlined above, The Little Database departs from these in its play through a focused exploration of an intertwined nexus of independent platforms for transcoded artifacts. The book is organized into four chapters, each addressing a primary little database and a correspondent methodological framework and file format. In an extreme of abstraction, these can be enumerated as follows:

1. computation, text files, and Textz.com;
2. preservation, image files, and Eclipse;
3. transmission, sound files, and PennSound;
4. dispersion, movie files, and UbuWeb.

This study proceeds from the macroscopic to the microscopic, and back again. It arrives at a close reading only in the final chapter. The first chapter attends to computation as the preeminent mode of critical data practices focused on parsing “big” data. The second turns to preservation, an issue at the core of archival questions within the digital humanities on a narrower band than computation at scale. The third examines the far-reaching circulation of sound files across multiple databases at the level of a small set of files rather than an archival collection. In the final chapter, I arrive at a close reading of three specific movies. From there, close readings expand into the broadest set of concerns, connecting back to standards and databases at scale. Operating kaleidoscopically, the closer we might get to a digital file, the more each pattern radiates outward.52 More specifically, the act of radiating outward may lead to expanding networks of standardizing protocols, data infrastructures, conditions of display, algorithmic bias, patterns of circulation, and the technical and cultural issues that by necessity remain only partial captures bound to the modest scale of any little database.

The first chapter, “Textwarez: The Executable Files of Textz.com,” examines computation through plain text standards and a series of playful visualizations and scripts deployed on a little database called Textz, edited by the German net artist Sebastian Lütgert from 2000 to 2004. I argue that the patterns of format analysis, container technologies, and computational performances that characterize the Textz project provide a basis for reading the increasingly complex formats examined in each following chapter. Ranging from experimental poetry and cyberpunk fiction to romantic literature and political tracts, the multilingual little database of Textz presents an idiosyncratic collection of text files that is at once too varied to compute and yet highly selective—not unlike a collection of files gathered over the years on any laptop. I contend that the text files hosted by the site shift into something more enigmatic once these texts are framed as a set of “textwarez,” a neologism marking the files on the site as “executable software” that operates on their human readers. I explore the text file as a sophisticated reading interface for human and nonhuman operators that emphasizes searchability, portability, and transmutability. Expanding on the formal properties of the text file, the chapter concludes by reading a set of applications released by Textz that call the stability of digital file formats into question through a playful performance of encoding and decoding protocols among a variety of media formats. In addition to revealing new dimensions of the text file itself, these applications offer a provocative statement on the politics of distribution, the interchangeability of formats, and the performance of computational scholarship at large.

The second chapter, “Distributing Services: The Periodical Preservation of Eclipse,” explores preservation and periodicity in the image-based Eclipse collection founded by Dworkin in 2003. Eclipse produces facsimile editions of rare and out-of-print works of experimental poetry and poetics, primarily originating in the 1970s and 1980s. Evaluating trends in periodical studies, this chapter considers Eclipse as a metaperiodical, with occasional releases of full magazines built into its archival database. To materialize this juncture, I offer a close reading of the formats that preserve L=A=N=G=U=A=G=E magazine in an extended comparison of a database of images, a set of reading copies, and the print run of the magazine. The highly compressed images hosted by Eclipse are seen to reveal unexpected layers of transcoding at play in the republication of this influential poetics magazine. From Xerox printing in the late seventies, to the GIF images presented on Eclipse, to the ongoing development of Adobe PDF capabilities, this decidedly low-fidelity magazine finds revised signification in its shifting formal and bibliographic codes. The distributing services of the website, I contend, paradoxically amplify the republication practices of the print magazine, while the site itself illustrates a resistance to the preservation systems of the internet. These resistances are demonstrated in an extended discussion of the archival captures compiled by the “Wayback Machine” web crawler developed by Alexa Internet for the Internet Archive. In contrast to the stripped-down text files of Textz.com, which emphasize algorithmic use, Eclipse presents transcoded images for human reading. Similarly, unlike Google Books, Eclipse remains mostly unsearchable, asking: what is lost and what remains of the book beyond the algorithm? Caught with losses alongside gains, this chapter concludes with a discussion of the stakes of compression, which carry throughout the book.

The third chapter, “Live Vinyl MP3: Echo Chambers Among the Little Databases,” opens by returning to “The Defective Record” by William Carlos Williams. Following the movement from gramophone to MP3 as detailed above, I turn to the same procedure’s reversal, tracking how a single glitched audio file produces a network of poetic practices centered on the vocal performance of Tracie Morris. These transmissions, often happenstance, capture the contingent forces that generate the meaningful interactions and poetic productions driven by a poetics of media formats. Amplifying the transformative properties of transmission, the chapter traces an autobiographical narrative of a small set of MP3 files recorded by queer Canadian poet, painter, and musician bill bissett in 1968. I track the passage of these files as they move through several little databases online, with a special emphasis on the PennSound collection. Founded by Charles Bernstein and Al Filreis in 2005, PennSound has digitized and hosted tens of thousands of poetry recordings, which have been freely downloaded by hundreds of millions of users. The narrative of this chapter tracks the dispersion of bissett’s album with Th Mandan Massacre, Awake in th Red Desert, through its initial release on Mutant Sounds, to its subsequent uploads to PennSound and UbuWeb. By following the thread of a particular poet’s output, in an attempt to tease out a description of each online collection, this chapter investigates how little databases inflect the works they host, and how we might begin to understand contemporary iterations of audio formats through this network. I argue that, while Mutant Sounds and PennSound host precisely the same MP3 files, the textual conditions—and thus the significance—of these two iterations could not be more radically different. By way of conclusion to this narrative, I examine a remixing tool entitled MUPS, developed by David (Jhave) Johnston, which introduces a new interface to a select set of files hosted by the PennSound archive.

A final chapter, “Dropping the Frame: From Film to Database,” examines the dispersion necessitated by close readings of media-specific films streaming in the UbuWeb collection. The UbuWeb platform, founded in 1996 by Kenneth Goldsmith and supported by a team of mostly student volunteers, hosts an extensive collection of text, image, sound, and “Film & Video” art. The proprietary nature, cultural dominance, and technical affordances of Adobe Flash once characterized the works hosted in this section of the site, and their traces remain long after the standard was discontinued in contemporary browsers. Reflecting on internet history and changing protocols for even the most pervasive file formats, this chapter reads through the ways in which media-specific film and video works amplify the effects of digitization. A close reading of Nam June Paik’s Zen For Film connects to the various remixes that artists have performed on the work to reveal the unpredictable layers of transcoding that Paik’s film releases online—all in stark contrast to the celluloid reflexivity of the original filmic performance. Similarly, once situated within the browser, Vito Acconci’s Theme Song shifts from the impersonal context of the art gallery to the intimate seduction of a confessional video blog. In the opposite direction, I also explore contemporary movies made specifically for internet release, honing in on a database movie by People Like Us (Vicki Bennett) entitled We Edit Life. By blending colorful pixels with aged frames drawn from found footage films, Bennett zooms in on the unwieldy distortions of digitization. On UbuWeb, these videos stream alongside the text, image, and sound formats examined in previous chapters. I conclude this reading with an extension to the little databases hosted on every computer, from iTunes music libraries to folders full of PDF files to increasingly uncommon hard drives packed with downloaded movies. These examples build on previous chapters to present an argument for a renewed attention to the poetics of the little database, attuned to the media formats that continue to shape user experience of art and literature online.

Finally, in a short conclusion, I discuss the relation of downloaded content to streaming media. Rather than displace downloads, the always-on streaming internet allows us to interrogate what it means to gather a collection. Indeed, a history of digital practice throws into relief the everyday repertoires we continue to employ. Today’s social media driven internet is built alongside the “web 1.0,” which persists in its shadow. I point to the Electronic Poetry Center (EPC) as a particularly interesting example of happenstance longevity. Founded by Loss Peqeño Glazier with Charles Bernstein at the Poetics Program of the State University of New York at Buffalo in 1995, for a very brief window, the EPC was the largest site on the public internet. Now archived and maintained by the University of Pennsylvania, the site is full of holes, anachronisms, and broken links. Within these glitches, an alternate vision of the network continues to subsist alongside mainstream practices. Here, the little database presents an opportunity to reimagine the internet a little more strangely, where future developments might rediscover older forms of invention made long before the establishment of the generic conventions that undergird the 2020s internet as of this writing.

The media poetics of these chapters carry over into a series of interludes that outline practical methods to perform the arguments of The Little Database in variable formats of creative scholarship. Presenting reflexive performances of each chapter’s argument, these interludes articulate the project in their own terms, through medial formations that the written document cannot address. My approach to these interventions is inspired by a practice of creative remaking articulated in the essay “Deformance and Interpretation” by Jerome McGann and Lisa Samuels. Calling for “a practice of everyday imaginative life,” they offer a set of tactics for creative scholarship as a way to open up the conventions of academic prose to experimental discovery.53 Following this line of inquiry, each of the four interludes in this book seeks a reading of the little database from beyond what McGann and Samuels call “the textual looking glass.”54 It should be noted that these interludes do not function simply as supplements; nor are they merely additional creative works that accompany discrete chapters. Just as Bernstein’s definition of poetics includes the “continuation of poetry by other means,” this project performs the media poetics described here as the continuation of scholarship by other means.55

In this respect, the creative work discussed in each interlude could effectively stand in for corresponding chapters of the book. That is to say, this project does not simply address the little database, or inscribe the ways in which works of art and literature are transformed by digital formats—it also demonstrates how scholarly performances of variable formats might occur through creative methods of media poetics. Of course, depending on how the reader might find these words, this practice is variously determined by the affordances of the black-and-white print codex, the full-color ebook, the open-access digital interface, or hopefully, the rogue PDF file nested within a shadow library. In print-based editions, the reader will find typographic links to expanded digital materials for all creative interludes via the project’s open source Manifold Scholarship page (https://manifold.umn.edu/projects/the-little-database).56 The materials I present offer another way to reveal the source code behind the arguments I make while also amplifying the play of meaning within the transformative media formats I discuss throughout. These mediatic performances, once more, are presented in every instance as a model for yet unimagined interventions that the reader might make.

This book is an invitation to play with the little database as one might piece together a slideshow for loved ones, produce a mix-tape for a significant other, or even capture citations for an essay. It aims to open up personal collections to the same methods that might unpack an institutional archive. To offer a range of tactics to read cultural history through playful engagements with the contingent networks of the present. It hopes to open new windows for the reader to explore undiscovered networks of meaningful experience among experimental arts, internet cultures, ordinary collections, and the play of variable media formats.