A Trans Historiography of Glitches and Errors Free

What does it mean to write a trans historiography of glitches and errors? I write this history with the awareness of the embodied relationship between trans life and institutions, taking from trans studies the constant political awareness of the relationship between trans life and systems of governance enacted through nations, states, and institutions, which have become invisibly part of the binary rule sets enacted within computer systems.

Susan Stryker, Paisley Currah, and Lisa Jean Moore write: “‘Trans-’…becomes the capillary space of connection and circulation between the macro- and micro-political registers through which the lives of bodies become enmeshed in the lives of nations, states, and capital-formations, while ‘-gender’ becomes one of several set of variable techniques or temporal practices (such as race or class) through which bodies are made to live.”⁴ The micro-political acts and practices of trans bodies are deeply affected by the macro-political regulation of gender and trans life. Trans is necessary to understanding this media history, and necessary to understanding Fenton and the glitch. Trans history is itself a history of mediation: it is tied directly to histories of trans people told through documentation, bureaucracy, surveillance, archives, and other forms of mediating power of the state and nation. Stryker writes:

Trans history is a history that has been told through others, through documents, through institutions, through mediation. To write a trans media history is to be aware of what it means for trans lives to be taken, represented, and told to others, and the history of these practices. A trans media historiography must reconsider what mediation is, and what mediation does, and ask: How can one write a trans media history when mediation itself has such a violent history in relationship to trans people and the histories we are told through?

What does it mean to write trans media history as a history of unmediation—that is, a history of undoing mediation? A history of things that cannot be documented or mediated, or things that evade or dismantle mediation in relationship to trans life? Unmediation can take several forms. The first is considering trans media history as one that evades knowability, with trans life as “missing” from representation, archives, and recording within histories of media (and particularly computational media). When writing trans media history, we must always be aware of that which cannot be saved, that which will always evade documentation and mediation, and things that have not been recorded or cannot be recorded in relationship to media.⁶ Secondly, unmediation brings to the foreground the construction and interruption of mediation itself as an active and often linear process, a movement of a signal from sign to signified, input to output. Here, unmediation is the awareness of how power circulates within and around media technologies, and the undoing of these systems. With regard to Fenton, unmediation takes the form of the breakdown of the screen through the glitch, which makes the user aware of the construction of the computer system, and the user’s own interpellation (or lack of interpellation) within these systems.

Interpellation is a word that can be used to describe the way users and citizens are addressed and the way users and citizens identify themselves within systems of governance. Wendy Hui Kyong Chun discusses this in relationship to software, writing that “importantly, the ‘choices’ operating systems offer limit the visible and the invisible, the imaginable and the unimaginable.…You are not, however, aware of software’s constant constriction and interpellation (also known as user-friendliness) unless you find yourself frustrated with its defaults.”⁷ Chun draws parallels between software and citizenship and the ways in which institutional systems of governance and computer systems both interpellate or expect and anticipate the actions and presences of certain users, certain citizens.

The glitch itself poses a momentary experience of undoing, unmending: the artifacts that appear on-screen show us the breakdown of these systems and a fleeting awareness of the way we are imbricated, or unable to be imbricated, within its systems and structures. The glitch might appear as blocky artifacts that take the shape of ghostly faces when streaming a low-latency video online; the crashing of Microsoft Word or Photoshop when it is unable to complete a task; or a moment in a video game when an avatar’s body sinks through the ground, clipping through the boundaries of the world itself. These everyday glitches—happenstance, unintended, random—call attention to the fact that these software systems are designed to be immediate but in reality are mediating, serving as an intermediary. To unmediate is to call attention to this continuous mediation, to the continuous interpellation we experience through media, to the fact that systems of governance function similarly—that some people are interpellated fully into systems of power, while others are not.

For many trans people, and especially Black and Indigenous trans people and trans people of color, this experience of interpellation is broken constantly, revealing the ways in which government systems do not anticipate our presences. We become aware that there are many mediating steps that comprise seemingly fluid, everyday actions like moving through workplaces, homes, and public spaces, or getting access to passports, visas, and medical care. To unmediate is to see the constant systems of rules, the violence of representation and documentation, and the ways in which mediation does not work in expected ways, too, for trans people, or for trans history.⁸

The glitch itself is revelatory in relationship to power structures. While some might experience glitches as accidental moments in their everyday use of technology, such moments are still moments of undoing. In the case of Fenton, it is the pointed and continuous undoing within Digital TV Dinner that allows for a startling suspension of this glitch moment into moments, and these moments into minutes. This suspension of the glitch undoes the computer, the game console, and the screen almost to the point of complete illegibility. As we watch Digital TV Dinner, we are resting in a moment, expanded, in which we are no longer interpellated, in which the screen itself is no longer functioning as a representational intermediary, in which we watch computational interfaces dissolve from symbols to signals that blink, irregularly, struggling to signify.

In positioning the glitch within this trans historiography of glitches and errors, we must consider the tactic of suspension. While the glitch itself is an everyday, and oftentimes accidental, moment in relationship to media technologies experienced by many, Fenton’s experimentation with duration, length, and suspension play a part here in positioning the glitch as a trans mode of production. The glitch, as mobilized by Fenton, is not a fleeting moment of failure, but two minutes and forty-one seconds of undoing. It is a collection of the effects of the repeated bodily acts that Fenton imposes upon the machine. It is a constellation of moments, dotted through time, spliced and edited together in the glitch video we know as Digital TV Dinner. The video itself is a sustained and lingering deconstruction of the sovereignty of the computer system. It highlights the ways in which we might imagine alternative possibilities, and alternative futures, for the computational rule sets we have internalized so greatly that they have become invisible.

I write a trans historiography of glitches and errors with the awareness that history itself is a form of mediation: an assemblage of information that takes the shape of a person or objects that are deemed meaningful or important. Fenton was the “first” in several domains of history, but it is not enough to position transgender life in histories of digital media as a series of elided “firsts.” To write a trans media history, we must also consider unmediation: the undoing of the way history mediates, with an awareness of historiography itself as a set of practices that hierarchize certain modes of knowing, of being, over others.

A “traditional” history of Fenton and her place in computing history might be positioned with regard to industrial histories of games and computation, focused on dates, objects, and firsts: The fact that Fenton was one of the founders of MacroMind, a software company that operated in Chicago in the 1980s, which later became Macromedia and was then acquired by Adobe Inc., which produces software like Adobe Photoshop and Illustrator. The fact that Macromedia Director, a program Fenton created, became Adobe Director, which was the precursor for what we know as Flash, a platform used in most major web browsers to run games and interactive and visual media.⁹ The fact that she led the team that designed and implemented the Bally Astrocade home computer and video game console, which at $300 was the first and cheapest of its kind for six months.¹⁰ The fact that the Astrocade utilized the Zilog Z80 microprocessor and therefore situates trans life, labor, and history within the movement of computers and microprocessors from institutions to home spaces.¹¹ We might also consider the history of Digital TV Dinner’s debut at the University of Illinois at Chicago’s Electronic Visualization Event 3 in 1978 and its later broadcast on WTTW’s Image Union public television program in 1979.¹² Or that the Bally BASIC programming language she adapted for the Bally Astrocade home computer and game console was designed to be sold to households with coding literacy in mind. And lastly, the economic and capitalistic failure of the Astrocade in the wake of the Video Game Crash of 1983 and how this movement toward “accessible” home computing in terms of both the cost of the computer, and the use of the computer as a learning tool for programming, might be viewed as one possible trajectory of many for the home computer that was later outmoded and overcome by companies like IBM, Microsoft, and Apple.

To write a trans historiography of glitches and errors is to unmediate an entire mode we use to write and tell media histories. Noting the “firsts” or “onlys” of something, especially with regard to queer and trans histories of digital media, is itself an eliding process, which reinforces the idea that existing histories of technology are already complete timelines with momentary lapses in acknowledging the contributions of marginalized people.¹³ Many existing histories of technology are centered on computational objects and their preservation, which is predicated on the institutional form of the archive itself and its systems of funding, ownership, physical space limitations, and relationship to institutional modes of deeming who matters, what matters, and what is saved.

The “addendum” and “revision” are regulatory and necropolitical acts that have been enacted in a number of historical and institutional circumstances with regard to queer and trans life, and especially Black and Indigenous queer and trans lives and queer and trans lives of color. To write history is to produce an assemblage of facts and information, and this history is highly mediated, historically situated, and revisionist, not unlike the Diagnostic and Statistical Manual of Mental Disorders itself as a clinical archive that documents the changing orientation of the American Psychiatric Association, and the US medical-industrial complex at large, toward the regulation of gender and sexuality.¹⁴ The revision and the addendum are institutional acts of knowledge curation that reinforce historical systems of governance with regard to who may revise these histories, why, when, and for what reason. Achille Mbembe writes in “The Power of the Archive and Its Limits” (2002):

We must consider the way that histories of technology have been written as careful collections of documents and facts narrativized into movements toward technological “progress,” which is itself a racist and colonialist ideology. We must also call attention to the methodology of “adding back” elided marginalized people to histories of games and technology as, in and of itself, a historiographic process that has the potential to reinforce the same systems of power that elided us to begin with. The “adding on” of marginalized people, especially queer and trans people and people of color, to these supposedly linear histories (which are overwhelmingly focused on dates, objects, firsts, financial successes, and market saturation) reinforces existing modes of historiography: who we deem important, why something is important, what people thought was worth being saved, and what space they had available to save it. Laine Nooney criticizes the addendum as a historiographic method in relationship to women in video game history in “A Pedestal, A Table, A Love Letter: Archaeologies of Gender in Videogame History” (2013):

To write a trans media history, we must also be aware of the mediation of history itself, and call attention to the modes with which facts, dates, archives, and data have stood in, mediated, and become representations of trans life and trans history. We must be aware of media, and the act of mediation itself, as always necessarily materially and historically specific, imbued with their own practices and politics. I write this trans historiography of glitches and errors with the awareness of what has not been saved. Re-centering the ephemeral, and the acts, erasure, and embodiment within computer and video game history, allows us to argue that what has been lost—the objects missing from archives, the trans histories that are left out of historical narratives, all things that have not been recorded—have a life and a history that we can still trace, and which is still meaningful, that still anchors us as trans people to our pasts, presents, and futures.

The history of computational media is a history of enumeration—of “computing,” the word taken at its most basic value—and the use of computers to hold information meant to stand in for lives, time, and locations, rendering them as data. Populations are accounted for through the mapping and accounting of bodies to smaller and smaller units that could fit into smaller and smaller spaces: the punch card, the transistor, the microchip. We might read the history of computing as a history of density, a history of compression, a history of space, motion, of inside and outside.¹⁷ In situating the body as documented inside of these systems, I also ask, is there an outside to this system? And what might a trans history of glitch tell us about imagining the possibility of an outside to these technologies of sovereignty?

Nael Bhanji writes about border crossing, motion, and movement, noting that narratives of linear progression that establish before and after and inside and outside as fixed objects, tied to place, are colonialist frameworks for orienting narratives of trans embodiment and history:

I return here to the idea of an outside and an inside to computer systems, their coded enforcement of rule sets and systems thinking, and how we might imagine the momentary glitch itself as this “non-zone”: a place of movement, fluctuation, and in-between-ness that calls attention to the fact that the glitch is a motion, a movement, a flux rather than a resting place, a point A or a point B. The glitch presents us with moments of fleeting possibility and promise—imagery that signals to us that there is something beyond the coded boundaries of rule sets within software systems, that there might be the possibility for something unintended, something unstable, something unimaginable whose status is never quite fixed, never quite bounded.

The processes that computer systems are purposefully coded to perform might be viewed as an inside to the system, with a purposefully programmed and known set of outcomes. For instance, if a user types a command, the computer is programmed to either accept the command or reject it due to not being able to process it, and the system will reveal this error to the user with an error message, or the simple inability to carry forth with a task. This way of conceptualizing software error does not break the computer. Instead the computer is able to enforce its limitations and regulations on the user—“No, I can’t compute this information, it’s not numerically possible,” the computer might think, and the computer might respond to the command to divide two by zero with an error message in the visual interface. The software error internal to the computer system is designed to communicate to the user the boundaries of what the computer can do, effectively enforcing a limitation on the user’s ability to manipulate the system before the user can exceed these boundaries and break them.

If we conceptualize the error as inside, we might be able to position the glitch, a kind of software failure, as operating outside—a breaking and overcoming of the computer system and its coded boundaries, a melding and dismantling of these rule sets as sovereign. In the glitch, failure is positioned not as a state but an act that dissolves these boundaries, breaks these limitations, and in turn produces completely uncoded, unseen, and unanticipated visualizations within the computer’s interface.

The Bally Astrocade relies on a programming language that Fenton wrote called Bally BASIC, or Beginners’ All-Purpose Symbolic Instruction Code, which was based on Li-Chen Wang’s programming language Palo Alto Tiny BASIC. The 1981 version of the instruction manual for Bally BASIC begins with the following:

Marketing flyers from 1979 called the console the Bally Computer System and noted that it contained an “exclusive, built-in Bally Brain,^TM a powerful microprocessor, [that] creates a ‘memory bank’ of more than 12,000 instructions” (fig. 6).²¹ The cassettes (also called cartridges) sold for the Astrocade were unique in that they also contained their own chips for memory storage, where users could “add up to 8,000 more bytes with each Videocade^TM Cassette.”²² Much of the language in these marketing materials positioned the Astrocade as a computer that would allow users to become familiar with coding and creating visual media, including games, digital art, and computer-generated audio, programmed using Bally BASIC, which came packaged with the Astrocade as a separate Videocade cassette. “It’s instant creativity with Bally BASIC,” a second advertising flyer reads. “This programming cassette is an easy-to-understand version of the BASIC computer language [used] to help users create their own games, art, and electronic music. With these important exclusives: A self-teaching course that lets users create programs in minutes, an easy to use programming keypad.”²³

A 1981 marketing flyer highlights the bundling of the Bally BASIC cartridge with each console as one of the console’s main selling points: “The most fun game of all is learning to create your own computer programs. Now there’s no easier way to learn about computers than with the new Bally BASIC system. This plug-in cartridge with built-in audio tape interface converts the ARCADE into a personal computer you can program yourself” (fig. 7).²⁴ Fenton and the Astrocade’s creators imagined the home video game console as a full-fledged computer that could code and produce its own software. The manual for the Bally BASIC, which she wrote with Dick Ainsworth, George Moses, and Brett Bilbrey, includes lessons on “Printing, Counting, and Loops,” “Subroutines,” and “Arrays.” After these primers, the manual then moves on to lessons about writing code to produce computer-generated media with sections titled “Video Art,” “Electronic Music,” “Graphics,” and “Computer Games” (fig. 8).

In a 1978 paper given by Fenton, Thomas A. DeFanti, and Nola Donato titled “BASIC Zgrass: A Sophisticated Graphics Language for the Bally Home Library Computer,” they outline their vision for making the programming of computer graphics and computer-generated art accessible to home computer users. They situate the video game console and computer hybrid within a particular historical moment for home technology users and hobbyists, describing the movement of computers to home spaces within the context of amateur (or ham) radio and the computer literacies required to teach users to code:

Fenton, DeFanti, and Donato position the Bally Astrocade as counter to the computer hobbyist movement, writing that ham radio was saturated with home users who were experts, well-versed in the technologies they were using from home.²⁶ They write that minicomputers, which were very recently becoming household items, were similarly objects of technical proficiency, rendered inaccessible not so much due to cost, but due to lack of access to household programming literacy.

The error message is positioned within this paper as part and parcel of computer literacy and as a disciplinary method to structure and regulate user interaction. The manual for Bally BASIC discusses how errors are communicated to the user, with advice like, “If you make a mistake in punctuation (as in leaving out a comma), the computer cannot run your instruction. If this happens the computer will print the instruction on the screen with a question mark in the position of your error, to show you where your mistake is.”²⁷ In the same conference paper, Fenton, DeFanti and Donato propose the programmed software error as a kind of pedagogical tool that would allow coding and programming to become easier for novice programmers:

The Bally BASIC programming language is built around simplifying the signaling of error on the behalf of the computer to the user. The error is a system that has always been in place—the computer signaling in various capacities that the user’s input is illegible, incomputable, unrecognizable, unactionable. That the user’s capacity for acting within the system is not possible.

***

In early 2020, before the shutdowns due to COVID-19, I made several research trips to the Phil Morton Memorial Research Archive, housed in the School of the Art Institute of Chicago. The collection was curated by Professor Jon Cates, and I visited under the supervision of educator and glitch artist James Connolly. The archive itself is a room filled wall-to-wall with old magnetic tapes in a variety of formats, from Sony U-matic to VHS. Over the course of several weeks, I digitized a small group of U-matic tapes that, I had been told, had never been viewed or dubbed before. Among the videos I digitized was a series of KC-60 U-matic cassettes with the titles “B-BASIC WITH DAN” scrawled on them in block letters and numbered from one to three (fig. 9). The tapes begin with a video capture of a television screen hooked up to the Bally Astrocade. The voice of Dan Sandin, a computer graphics artist and creator of the Sandin Image Processor (an analog synthesizer that was used to augment video art in real time) narrates the code he enters and executes on the screen.²⁹ Throughout these three tapes, Sandin gives a live demonstration of how to program sounds and images with the Bally Astrocade home computer.

Sandin’s first demonstration on the first tape is a lesson on the error messages that have been programmed into Bally BASIC programming language. Sandin begins, “Let’s start with the error messages.…There are three error messages.…Now, a normal large computing system normally has several hundred different error messages. This one, which is a very minimal system, has three.”³⁰ The three error messages, as Sandin demonstrates, are programed to be presented by the computer to the user when the computer does not understand the user’s input: WHAT?, HOW?, and SORRY. These error messages are earnestly colloquial, each signaling a different type of lack of comprehension on behalf of the computer and the limitations of the user. In the video, Sandin goes through the process of manufacturing circumstances in which these three error message would appear on-screen.

WHAT? appears when the computer cannot recognize a statement. The example he gives is a statement with incorrect syntax—he types >PRINT DAN, which the computer cannot take in as a list of commands, asking WHAT? Sandin then corrects the statement to >PRINT “DAN”, with quotation marks, which allows the computer to execute the statement and print the word “DAN” on the screen (fig. 10).

HOW? appears when the computer understands the syntax of the statement but can’t execute the statement. In this case, Sandin asks the computer to divide ten by zero. He types >PRINT 10÷0, and the computer responds with HOW? He then corrects the query to >PRINT 10÷1, and the answer is printed: 10 (fig. 11).

SORRY appears when the computer does not have the hardware capability, specifically memory in this case, to execute a process or command. The example Sandin gives is >PRINT @(1000), to which the computer responds with SORRY (fig. 12). Sandin explains: “What happened here was that it tried to find the value stored in an array number 1000. It’s in a sense a number one bigger than I’ve got…available memory for, and the ‘sorry’ command is an error message that says ‘Sorry, I’m all filled up, you have to delete something.’”³¹

There is a cybernetic loop that the user is encouraged to perform with the computer through these error messages—a constant, preprogrammed stream of feedback between user and machine where the user interacts with the computer in ways the computer anticipates (typing out commands using the Bally Astrocade’s keypad), and the computer announces to the user the possibilities of what is programmable, and what is error, through its user interface. The visual interface here functions disciplinarily: it encourages the user to continue interaction with the computer within the terms that the computer communicates to the user. In Bally BASIC, the error message is designed to be corrective: it signals to the user what is possible and impossible for the user to do, and the user learns from this what actions are possible and encouraged and which are not, enforcing a feedback loop governed by the software system that dictates the behavior of the user.

The error, manifested in the screen, is a signal. These signals manifest in the Bally Astrocade as blank, uniformly colored backgrounds with black text. They manifest as letters that form legible words and syntax, communicating to the user the limits of what is possible and able to be executed. Fenton’s coding of the hardware limitations of the computer through the error message apologetically, and colloquially, reinforces a coded boundary that has been positioned to prevent the user from going any further, from affecting anything more, that signals the end of motion or possibility. SORRY, the program says. SORRY.

In 2018, Fenton created a video titled Primordial Glitch Art. In it, she sits in what looks like her home office. Behind her on the desk is a trapezoidal box with a wood-grain pattern adorning its front lip and sides: the Bally Astrocade. This Astrocade is missing the smoky plastic lid that normally sits on top of the console, revealing a recess filled with an array of cartridges. From my own experience encountering the console at the Phil Morton Memorial Research Archive at SAIC and the Strong National Museum of Play in Rochester, New York, I know a few additional things, too: That the console is largely covered in dark brown lightly pebbled plastic, with wood veneer covering its sides, and yet it still has a heft and weight to it. That the gold trim lining the edges of the console is gold paint on plastic that flakes off over time, but the golden number pad has a thin aluminum-like covering that is cold to the touch. That the console very easily accumulates dust in its crevices and corners. That the Astrocade is very prone to glitches: when I first turned on the console and inserted a cartridge at the Strong, I was met immediately with a screen filled with bright repetitions of pink and white horizontal lines that appeared to break and rejoin in jagged steps from the upper left to the lower left of the screen (fig. 13), an image that persisted until I turned the console off.

In the video, Fenton faces the camera, with long salt-and-pepper hair tucked into the neck of her gray T-shirt. “What I’m using here is a Bally Astrocade computer hooked up to a TV to channel 3,” she says. “What I do is I take one of these cartridges, I insert it into a slot here.” She holds a copy of a cartridge titled Blackjack (1978) and presses it into the console. “I press reset. Then I usually enter into the cartridge’s thing, and I spit the cartridge out.” She starts up Blackjack and arrives at its bright green opening menu, which reads “Player 1 Place Bet.” Once the cartridge is spit out, this menu flashes into a series of black vertical lines of different widths over a white background. This screen flashes to this pattern for half a second before the console resumes normal function and returns to the Bally’s “SELECT GAME” menu. “And you see, you get a glitch,” Fenton says. “Let’s do a few more of these. Depending on how lucky you are, you get different types of glitches.”

Fenton describes out loud what she is doing with her body, and how the console is responding to her. Because the generation of these glitches is so repetitive, she describes the same procedure several times as she executes the same actions over and over throughout the one-minute, forty-three-second video. “I work the user interface here, make some selection, and at some point while the code is actually executing the game, I spit the cartridge out and press the eject button,” she says while producing another glitch. This time the screen flashes briefly between a green screen with blocky pixels at the bottom (fig. 14) to another frame filled with vertical lines of varying widths (fig. 15). Each frame lasts barely a second, maybe even a fraction of a second.

The act of pressing the eject button is something she performs using both hands, with one hand operating the Astrocade’s keypad and the other pressing the eject button under the cartridge slot, with one act happening quickly after another. The screen abruptly flashes to a barely half-second-long frame of a garbled image before quickly shifting back to the console menu screen. “And see, different types of glitches, different circumstances,” she says, as she produces another half-second error. She leans her body in toward the console. “Sometimes it resets just like that, other times…[something] more interesting.”

Fenton ejects Blackjack and inserts a Star Battle (1978) cartridge. Immediately the small TV screen fills with tumbling, serrated lines of pixels that cascade in intervals down the screen, like a Slinky falling down the stairs (fig. 16). The screen then washes from top to bottom in waves, producing long vertical stripes that look like a smearing bar code, a waterfall (fig. 17). The screen moves back and forth between these two types of glitches—the waterfall and the cascade—for twenty seconds. It is an impossibly long time considering the length of the previous glitches, which flitted in and out of existence. Here, the glitch lingers. We cannot miss it, we cannot mistake it for anything other than what it is. I watch the back of Fenton’s head as she gazes at the screen, watching the pixels move. She’s silent for a few moments. “Not sure what we did to deserve that one,” she suddenly says, glancing at the camera, smiling, then turns quickly back to the screen to watch. “That one just goes on and on and on,” she laughs.

***

When I watch Digital TV Dinner, all I can think about are the things that are not there, that are not represented, and the things that the video is not a record of: The labor of Fenton’s hands repeatedly jostling the hardware console, ejecting the cartridges, over and over, if only for a momentary, half-second cascade of pixels and blocks to appear on-screen. The time it took to generate enough half-second to ten-second glitches to fill a two-minute, forty-one-second video. The fact that Digital TV Dinner has been edited to isolate the glitches on the screen, and how these cuts remove the full duration of time it took to generate these images. I think of the home spaces in which this computer and video game console existed, and the spaces in which this piece of glitch art was created, and the bodies that necessarily must have moved, must have circulated around it, to produce these glitched signals. All of these things are not represented in the glitches on the screen, but these things must have happened, must have been there, for this video to exist.

The glitches in this video call attention to everything unseen, unsaved, and the undoing of the programmatic, the symbolic: the movement of sign → signified. The pixels in Digital TV Dinner assemble themselves in shapes of varying sizes and repeat in varying intervals, with configurations shifting from moment to moment. Some moments the interface dissolves into what appears to be a randomized assortment of small squares, dancing and moving across the screen like static or noise on a television set. Other times these blocks appear in larger sizes, with more obvious assemblages of repetitions and randomizations. In one moment, the patterns repeat in horizontal layers, pixels forming what looks like the mathematical curling of a sea wave or the curve of an Ionian column (fig. 18). In the next, the screen fills with vertical repetitions that take the shape of what looks like a series of spectrograms, or sound waves, turned on their side, reminiscent of the ridges and repetitions of a tapestry (fig. 19).

In all of this, there is constant motion: The prickling of static dancing across the bottom of the screen. The screen wipes that move downward and upward that clear the pixels from the screen in waves, or introduce new patterns of pixels. The constant and irregular leaps between one panel of pixelated patterns and another, with one image jumping from what looks like the ridges of a bar code (fig. 20) to patterns that repeat diagonally, like the topstitching on a quilt (fig. 21). There is a constant undoing and remixing of familiar assemblages: the top of the screen, reserved for titles and contextual menu information in the Bally Astrocade’s interface, is rendered into pixelated blocks or glitch-ridden text that scrambles, undoing itself. It begins with “PLAY A GAME” (fig. 22), later flashing to “PLAY A GA7” (fig. 23), and even later, flashing into blocks of scrambled pixels that take the shape and form of letters, words, and language, without connection to a signifier (fig. 24).

I see, in these moments, a computational system that cannot fulfill the functions it was designed for. Instead of the seamless procession of input → output, there is a breaking of this programmatic linearity, with the assumption that the screen fulfills a function, conveys a message, to the user. Instead with the glitch, the → of this equation stands on its own: a record of movement, of motion, of processing, of failure as a verb rather than a state. And here, in Digital TV Dinner, we linger in this moment, expanded into minutes, experiencing waves of this →, this motion, instead of lingering at the moment when the glitch began, and wondering when the glitch will end. We are affected by cascades of these glitches that seem to go on endlessly. In Digital TV Dinner the glitch becomes a wading pool, a “place to call home,” as Bhanji writes. We linger here, we are suspended here, in these moments, stitched together, losing track of where and when something begins, and where and when something ends. We sit here, in this moment of undoing, watching the structures and systems we know become continuously undone, looping, repeating, watching as shadows of the familiar interface are broken into components, into pixels that shudder on and off in patterns and arrays.

This is unmediation: undoing as a verb, a process. It is the lingering on the movement of a signal as it is translated from one state to another: sign → signified, input → output, with the → itself as an unquestioned and invisible intermediary, a type of mediation that moves a signal between positions. Unmediation is the questioning of the linearity of these processes and movements. Unmediation is a signal that is jammed, reversed, or broken, or called attention to and extended as it moves between states.

This, too, is trans methodology. Fenton’s patient dismantling. Her solidity, returning time and time again to the same seemingly unyielding systems. Her certainty that embodying the same actions, enacting the same movements, will produce the flickering possibility of different outcomes, unanticipated futures. Her knowledge of a system that is so complete that she knew there were more possibilities beyond the SORRYs, the WHATs, the HOWs. Her desire to linger in disparate moments of disruption, stitched together as evidence of the trans life that must have breathed, moved, labored, for something like this to exist.³² She shows us that the coherence, monotony, and sovereignty of the computer system does not and cannot dictate all that is possible, all that is available to us.

Two words in particular that Fenton uses to describe the process of producing glitches and glitch art strike me: lucky, and deserving. “Not sure what we did to deserve that one,” she says, as if the glitches are a gift we have been given, each time she restarts the console. Digital TV Dinner can only exist because of her patiently repeating the same steps over and over. As the programmer of the Bally Astrocade, she coded all of the software’s rulesets and boundaries. She knew what could be done with the console, and in silhouette, what could not be done. And yet there is a joy she takes in repeatedly running up to the same threshold and knocking at the same door, each time with both the hope and certainty of the computer producing something even she could have never intended or imagined. Her wonderment is startling: each time a glitch appears, she remarks that we might not be deserving of these moments of breakage, these instants of something else, something other that gives us a brief and wonderful glimpse beyond the possibilities tightly held and predetermined by a binary computer system.