Font of Possibilities

A durational performance for .otf files by Bruno Gola.
Starting on 21 of November 2024 performed by memesque.otf (current size is 8634KB).

Introduction

"This font was about to be thrown in the trash bin, but an invitation for an art upcycling project called “Art Swap” (Prater Galerie) gave a second chance to the file. Memesque is a fork of [Epilogue](https://fonts.google.com/specimen/Epilogue) (2020, Tyler Finck, ETC) by [Raphaël Bastide](https://raphaelbastide.com/) using [Skelefont](https://gitlab.com/edi8th/skelefont). It was generated as a draft in May 2024 for a project that eventually never happened. The font has been chosen for Art Swap by Raphaël Bastide for the exceptional accumulation of “default” compared to what a “good” font is supposed to be. Indeed, the font is overly bold, with bad contrasts, its glyphs are cropped, it has a very bad kerning, and suffer from many other graphic inconsistencies. Nevertheless, it exists, and can even be seen as a tribute to the contemporary meme imagery and typographic signature. I believe it can be usable in some contexts (no idea which ones).

Have fun with Memesque (life 2), that was a close one!"

Raphaël Bastide, in README.md

The Manager: [Turning to the CHARACTERS.] Who are you, please? What do you want?

The Father [coming forward a little, followed by the others who seem embarrassed]. As a matter of fact...we have come here in search of an author...

The Manager [half angry, half amazed]. An author? What author?

The Father. Any author, sir.

The Manager. But there’s no author here. We are not rehearsing a new piece.

The Step-Daughter [vivaciously]. So much the better, so much the better! We can be your new piece

Luigi Pirandello, Six Characters in Search of an Author

Some months back the invitation came for me to participate in Passing Data - Upcycling the Digital in the format of an art swap. My initial thought about an Art Swap was that, whatever file format that would come, I would like to explore its inner workings. I would not work with the file content as a computer user, but try to approach the structure of that file. Or how the computer and other software work with it. Kind of trying to go inside the computer and observe its world through its senses/ors, staying true to the digital material if that makes sense.

Back in 2016 during my application to study in the Berlin University of Arts, I remember writing that "code is my raw material". That was when I was first thinking about studying Art and Media, after years working as a programmer. However, now, while writing this text and thinking about it, I want to put it in another way. Code, as in “source code”, is one of the tools I use to work with the material. The material is the byte, and as bytes are structured, organized in order to obtain meaning to become file formats (and standards, protocols, data types, media, etc), and be rendered by software.

When I got Raphaël's files in a compressed .zip file, I unpacked it and quickly noticed the README.md file. It is a very common practice in Free Software and Open Source development to always include a README file. That is a text file that works as an introduction to that specific project, such as this introduction you are reading now. The author might write in the README file about the process, about how it came to life, the technologies involved, how to use it, what parameters to explore, where to start, etc. I started there, and while I read this README file, its first words already grabbed my full attention as I kept wondering about the computer trash bin and this font, represented by the memesque.otf file, sitting there, waiting for its final removal.

The trash bin was always a strange concept for me, and in fact I never used it. I always had a certain resistance to the office / desktop analogy in personal computers, so I used to just remove the file permanently without the trash bin period. The trash bin is the ultimate liminal space for bytes: files are put into the trash bin as a space in-between before they are completely erased, sometimes with an expiration date. I guess if the certainty of death means that one is alive, being thrown into the trash bin is a temporary proof that the file once existed before it is gone.

My idea with this project was to start from the trash bin, speculating what happens to the file once it starts rotting, decaying, losing information, until it's size is 0 bytes. I try to explore the origins of (digital) type, from movable type and the printing press, to the mechanization of the process, the jump from physical type objects to virtual layouts using optics and film and finally to its multiple digital forms, as bitmaps or bezier curves and different computer file standards. Here I am using the font as an object for investigation of the digital material.

Font files seem fascinating for that, they are very meta in their essence: since the 60s and the first computers operated with teletypewriters and early terminals we have been using text to replace the original binary instructions, those that were input to the computer via mechanical switches, punch cards or tape. The output of the computer usually includes some sort of text format as well. The computer needs some recipe of how to draw those texts, both while programming (input) or when displaying some result (output). Font files hold these recipes: how text, encoded as bytes, becomes readable to us.

This text you are reading now is static, as in a fixed-media work, but when rendered (when it is shown on the screen) in combination with the Memesque font it becomes an installation and durational performance, Memesque is slowly fading and losing its points, in a process that lasts 3 months (the duration of this exhibition). Glyphs become unreadable, the form is lost. The content however still stays there. Hidden in the source code of this page, rendered using yet another font if you open the source code inspector. Memesque will be gone, but in this process it gave origin to this work. And during this time working with Memesque it brought a multiplicity of thoughts and new realizations about the digital, about form and content, about materiality.

Matter isn't destroyed when decaying: it is broken down, converted into energy, releasing gasses into the atmosphere. It is fair to say that Memesque's process of decay worked in the same way for me, and hopefully for those reading this text. As if, instead of being thrown in the trash bin, it actually was composted, fulfilling its goal as a generative font, serving as a fertilizer for this and many new ideas, a source of possibilities.

Storage

The story of type is a story of storage, of our desire to record, transmit and replicate information. This function places type and fonts next to other technologies of memory such as oral traditions. But, while in oral traditions information relies on memory and performance (repetition of songs and stories through generations, for example), type embeds information into specific material, traditionally relying on physical materials like clay, paper, metal or stone. The way type represents and stores information is imposed by such material.

Using type to store information (as in writing) works as an interface between human thoughts and physical material, encoding knowledge into a form that doesn't necessarily rely on the human memory. Its longevity is connected to how durable a material is, and the access and understanding of the information is also completely tied to the choice of material and its physical properties.

In this sense, a font can be seen as a tool of interaction with stored knowledge. The arrangement of symbols (letters, digits, etc) tries to encode not just the words in a text but also its structure, rhythm, and emphasis: its intended meaning. Of course, as in any encoding-decoding process, essential parts of information might be lost.

It is clear to observe the process of decay and information loss in these early technologies of type and typefaces. Materials such as clay, paper, ink and metal are always prone to physical damage, to the effects of time and repeated use, corrosion, etc. Each material has its own characteristics.

At the same time, it is possible that such processes gave origin to new typefaces: a metal typeface that has been used for a long time might develop a specific texture. A broken typeface might reveal a new shape for a letter. This process could be seen not only as information loss but as a transformation of the material. Because of its interface nature, such transformations in the typeface also affect the meaning of a text, as when you change the font of a text from Arial to Comic Sans, for example.

Raphaël describes Memesque as a font that contains an "exceptional accumulation of “default” compared to what a “good” font is supposed to be. Indeed, the font is overly bold, with bad contrasts, its glyphs are cropped, it has a very bad kerning, and suffers from many other graphic inconsistencies.".

In the process of working with Memesque's material (points and curves encoded in bytes), I played with those features: I removed points, moved them around, replaced them, simulated physics, time passing, etc. Each of those transformations resulted in a new font, possibly giving new meaning to both Memesque and texts that would use Memesque.

Machines of Type

It is, as Braudel points out, “patient and monotonous efforts” which lead machinery on. Technical development is not only a matter of “the brisk changes we are a little too quick to label revolutions,” he writes, “but also the slow improvements in processes and tools . . . those innumerable actions which certainly have no innovating significance but which are the fruit of accumulating knowledge: the sailor fixing his ropes, the miner digging his gallery, the peasant behind his plough, the smith at his anvil.” These are the artisans, technicians, engineers whose work is more akin to “a collection of recipes drawn from craftsmen’s experience” than a tale of steady progress to some well-established end, and has “somehow or other evolved unhurriedly” by means of its own peculiar trials and errors, improvisations and accidents.

Sadie Plant, Zeros + Ones

With the Gutenberg press and mechanical reproduction, type foundries emerged to refine the craft of type design, manufacturing metal typefaces in different styles. This could be seen as the origins of many typefaces or fonts as we know today. At the time, letters were cast in metal as small rectangular blocks and arranged manually into words, a process that was very slow and laborious.

During industrialization, the rise of mass production demanded efficiency in typesetting. Linotype and Monotype enabled the automation of typesetting (combining typefaces into words), replacing the laborious manual arrangement of individual metal letters. Here, the "font" began to shift from a fixed, crafted object to something more dynamic: a set of instructions for creating forms repeatedly and reliably.

Both the Linotype and Monotype machines relied on matrices (as molds) for creating typefaces as instructed. Those matrices physically encoded the geometry of each letter, just like today a digital font file uses Bézier curves to encode the geometry of each letter. While automation brought a big shift, type still relied on physical objects made of iron. In terms of storage, we could say that each typeface was stored in metal.

In the mid-20th century phototypesetting eliminated the need for metal objects. Types were now defined by optical systems, and letters were encoded as patterns of light. Using photographic film, phototypesetting machines projected letterforms onto a photosensitive surface, allowing the original stored geometry to be scaled (using optics), and reproduced at different sizes.

With personal computers in the 70s and 80s came digital fonts. The initial digital fonts used bitmaps to store geometry. Bitmaps are grids of pixels, where each pixel is either 1 or 0 (on or off), just like a pixel drawing. This representation is resolution dependent: if you try to see a big bitmap on a low resolution screen, some pixels are left out and the shape might be different, and even become unreadable. Bitmaps were used because they are efficient to draw, it is a direct map from data stored as pixels to pixels to be displayed on the screen, but they are not very flexible.

With new computer-graphics technologies, new font formats were developed and vector-based fonts were adopted. Vector-based fonts store the information about the shape of a letter as Bézier curves. Stored mathematically, these curves can represent shapes independent of the resolution. As in vector-graphics, those shapes can be easily resized. That is a big shift, since until now fonts were always stored as "image", and now they were described as a series of mathematical instructions that the computer must execute before rendering to the screen.

Virtual

With the separation of the information encoding geometry and the material, as in phototypesetting, the font became virtual. The font was not a set of iron physical objects anymore, objects that one could handle with their hands. Information that described geometry of a letter was now a set of optical data stored on film or other optical medium. Phototypesetting marked the beginning of fonts as abstract systems that existed separately from their material manifestation. However, this virtuality was still tethered to physical media: film and chemicals, vulnerable to degradation and loss. It is possible to imagine the kinds of chemical and physical processes that could happen to the film that would alter and transform a font.

But, different from physical transformations on metal matrices, that had a direct relation on to the resulting shape of a letter, the effects of transformations on film would only be noticed when a phototypesetting machine was used to project a type. This marks the transition of fonts as virtual. Virtuality here resides in the font's potential. Its form existed in an abstract, intermediate state: not fully realized until it interacted with the photosensitive material.

Until the bitmap format, fonts were still stored as "images", as direct representation of shapes. With vector-based fonts (as PostScript) and Bézier curves, in the beginning of the 1980s typefaces were defined mathematically, independent of any specific medium. A font was now a set of instructions and it no longer depended on the constraints of the physical material as film or metal, or even the resolution or number of pixels in a screen. Its existence is now defined by the hardware and software that interprets its code.

This introduced problems when rendering, since the font might be shown in a pixel based display, the curves might not fit it perfectly. For that, modern font formats (such as Memesque's OpenType or OTF) have many features to help with the rendering process. Modern font files are small programs, full of instructions for the renderer (the piece of software software responsible for drawing the shapes described in the font file). These instructions are not only about each letterform, but also hinting about location, position, drawing style, and all that depending on the context (position in the text, neighboring letters, etc) of each letter.

A basic example of a feature supported by the OTF format is glyph substitution. If you write two letters together, the font might instruct the font renderer to substitute those two shapes by another one, usually so that it becomes more readable (think of languages that connect letters while writing, and such connection might change accordingly to each combination of letters). Memesque for some strange reason contains a rule that is applied when you use two letters "f" together as in "ff".

Those instructions are executed by the font renderer, just like a virtual machine runs programs written in an interpreted programming language. In this sense, the font file is much more than shape definitions, it is a source code.

While researching about this I came across a brillant talk by security researcher Julia Wolf describing a serious vulnerability in Microsoft Windows that was attacked by using a specific .ttf file. The attack explored exactly this state machine that runs code when certain characters are combined. The attack would be triggered by the combination of characters : (colon) and ) (close parenthesis), resulting in the classic ASCII smile.

:)

As you might have noticed, Memesque's definition of the colon character won't really show the smile as expected.

Digital decay

Digital decay happens in multiple layers. At the hardware level, objects like Hard drives and SSDs degrade over time, their electrons dissipating or mechanical components wearing out. At the software level, file formats can become inaccessible as software evolves and compatibility is lost, Operating System updates might turn entire libraries of files obsolete.

Unlike physical decay, where material degrades visibly over time, digital decay often operates invisibly. Bytes don’t “rot” in the same way organic matter does, but the underlying systems that structure and render those bytes are vulnerable to failures, obsolescence, and erasure. Think about when you try to open an old version of a project in a newer version of a Spreadsheet software, or when the Operating System gets an update and old software stops working. Digital decay manifests as corrupted files, obsolete formats, and inaccessible storage media. And it is even more perceptible in mobile devices because of their constant updates and black boxes.

Here, Memesque is playing a metaphor for this process, making a sort of digital decay visible: as it gradually loses its glyphs, its legibility is also gone. But the content of the text is still there. The decay doesn’t erase the text itself; it just disrupts the interpretation. What remains, hidden in the source code or within the file, is the residue of its original purpose, now unreadable in its intended form.

The possibility of accessing the source code in this case means that we can still retrieve the original content, we can still access the text and render it with another font. This highlights the importance of open standards and Free and Open Source software as a way of fighting against obsolescence, a path for a digital ecology.

The structure of a file, how bytes are ordered, serve as a map for software to interpret it. Without such software, the content is noise. Here, if the standard is documented, a programmer might be able to read the manual about how those bytes are organized and write a program to render its content in the way it was supposed to be. If such a manual is not available, the programmer can still try the process of reverse engineering: having a basic idea of how the final form should be, the programmer goes step by step investigating bytes, trying to make sense and find order in noise.

Obsolescence is, many times, a form of programmed decay that could be avoided.

Transformation

While the idea of decay in general evokes loss, in reality it is a process of transformation. In physics decay is the process of particles transforming into multiple other particles. In chemistry, as decomposition, it is the process in which matter is broken down into simpler forms. While, on a micro scale, in both cases this transformation seems to reduce the complexity of the original material, the process of decay can be seen as generative: it creates new material when observed from a macro perspective. As when a plant decomposes and its matter gets broken down into other organic material that is used as nutrients. It is an essential process of the nutrient cycle.

In the case of Memesque its decaying process is truly generative, it gave origin not only to this text and artwork, but also to simpler versions of itself. During the whole process of this performance new font files are generated, each "simpler" than the original Memesque, and at some point illegible, but still producing graphic shapes that might be used for other purposes than readable text.

More generally, even if for bad reasons (necessity, resource scarcity, closed source, proprietary standards, etc), digital decay is a fertile ground for creation. Many artists and Free Software developers find inspiration in so-called obsolete media, in reverse engineering of file formats that are not readable anymore or systems that stop working. System failures can invite experimentation and exploration, as seeing in the movement of glitch art and datamoshing, inspired by corrupted data.

In that sense, digital systems are a playground for working and thinking about decaying processes. Unlike rigid physical forms, the digital is inherently modular and iterative. Failures, glitches, obsolete formats and systems can often be easily recombined, recompiled and experimented with. Hopefully we can look at this process as decomposition, also part of a cycle where different digital matter clash and combine, to continue creating new works.

Possibilities

From those incontrovertible premises, the librarian deduced that the Library is "total"-perfect, complete, and whole-and that its bookshelves contain all possible combinations of the twenty-two orthographic symbols (a number which, though unimaginably vast, is not infinite)-that is, all that is able to be expressed, in every language.

Jorge Luis Borges, Library of Babel

In my "first language", Brazilian Portuguese, the word for font is "Fonte". Fonte is also the word for "source" and "fountain". The "font" has its roots in typography long before digital, it comes from the Old French word "fonte", meaning "something that has been melted" or "casting", a reference to the process used in early printing of melting metal into molds to cast individual letters. Like the internet meme says, "in Brazilian Portuguese we say Fonte for Source, Font and Fountain, and I think that's beautiful".

This process of investigation of digital decay made me think of the position of the artist as a "possibility space" traveler. The freedom and responsibility that speculation brings along, to go into weird directions that we have no idea where they might take us, until we get there. As a possibility space traveler we are allowed these explorations, and we can act on them. Artists must aim for expanding the borders of possibility. It is often thanks to the work of artists that we are capable of expanding our imagination.

In general I'm the opposite of a techno-optimistic, in fact I have a very critical view of technological development, tending to be skeptical of progress-at-any-cost. On the other hand, working on this project made me look optimistically at certain forms of digital decay as a source of inspiration, as a part of the creation cycle. For the past 20 years of my life I almost exclusively worked and used only Free and Open Source software, and now, after all this time, thinking about digital decay reminded me of the importance of open access, of Free and Open Source Software, of open standards, and also the importance of advocating for it. For that I am very grateful that Memesque, after escaping the trash bin, found me as the author to work on a new story.