Bug Report: Tuning to trans*feminist Xystem.Crash

on 31 October 2020 at ZAK Zitadelle Berlin with Katrin Köppert and with bug reports by Edna Bonhomme, Mario Guzman, Femke Snelting and Pinar Tuzcu

Engaging the bug report as a trans*feminist practice of affirmative critique that tunes to conceptual, political or material problems within existing systems. We invited writers to describe and problematize bugs at the intersection of computation, magic and decolonial theory. A bug report within computational practice is a document detailing what is observable about a piece of software that is not performing a dedicated task. In 1946, programmer Grace Hopper invented the term when operators found a moth in the Mark II computer. We engage the bug report as an anti-racist and trans*feminist practice of affirmative critique that tunes to conceptual, political or material problems within existing systems. Borrowing from Mozilla's bug report writing guidelines, we have distilled key trans*feminist debugging techniques towards naming problems within the processes that produce glitches, crashes and violences.

Join us for an evening of reading bug reports, discussing further implications and giving time and attention to collective bug finding on 31 October 2020 at Zitadelle Berlin

To Be Black, Free, and (En)raged A Film Collage by Edna Bonhomme

Flickering esoteric properties of executable code corrupts Full HD rendering of software culture potential by Mario Guzman

Summary:

Flickering esoteric properties of executable code corrupts Full HD rendering of software culture potential.

Component: (in which context does this problem exist?):

Executable code's flickering representation affects software development and its capability to be observed as an imaginative and critical thinking scenario. This perspective is visible through research on software culture (code studies) and digital artistic practices (creative coding, esolangs).

Version (historical background of the bug):

Software is not only the spinal cord of the technologies it enables, but all cultural practices involved in its development: interaction with political, economic, and socio-technical systems, development of creative solutions, metaphorization, writing, rewriting, configuring, hacking, erasing, debugging, etc. These practices also make up software, and consequently, software is both a semiotic and material/practical activity. These sets of practices did not start with the appearance of modern computers. In fact, the technical principle of computer software, controlling matter through manipulation of symbols, is also the technical principle of magic, understood as an esoteric practice (Cramer, 2005). Therefore, executable code possibilities were already available in spells, religious encoded texts, numerical musical composition, combinatory literature, and philosophy. As science rivaled and marginalized magic and religion, software development partially abandoned its correspondences between computation and its esoteric practices. For example, the harmonious link between macrocosm and microcosm, as Pythagoras understood it, or the divine mathematical correspondences of divinity, as performed in the Kabbalah. These developments were concealed in favor of using algorithms as efficiency, productivity enhancement, and exploitative tools.

Description including (larger detailed summary of the bug):

Computation’s history is full of nonexclusive contradictions. Algorithms can be perceived as an aesthetic material for poetic experimentation (Oulipo) or represent natural language acquisition parameters (Chomsky); computation can explore philosophical divination through randomized values (I Ching) or used as a statistical prediction development tool. These connections between computing and a wide range of speculative imagination embracing art, language, philosophy, religion, and magic are generally not recognized. On the contrary, the more restrictive elements of these mystical practices have embedded themselves into the industry's behavior and programming's social representation. Obscure writing, elitism, decreased accessibility, opaque inner workings only understood or intended to be understood by a small number of people with unique and secret knowledge. The initiated standing in an impenetrable realm. The chosen ones. The mysticism of programmers and their immutable machines. These representations are poured into our technology and end up configuring Artificial Intelligence' black boxes as a new kind of new demi-gods of logic, efficiency, and neutrality. But technology and algorithms are not neutral. Under this perspective, software culture's image appears to be flickering, oscillating between its enhanced imagination and its obscure practices. This bug corrupts software culture's full HD resolution, allowing low aspects ratios to render only its exploitative characteristics, instead of displaying the full spectrum of its explorative features. Nevertheless, this glitchy signal may show us that behind its secrecy, software is a cultural construct, an intellectual and social activity that enables us to produce knowledge through speculative imagination.

Additional Builds & Platforms (which other contexts are affected? what other positions does this have implications for?)

This bug skips version 19.70.s; 19.80.s. Popularization and accessibility to computers and programming boosted computation's cultural potential in the late 70s and early 80s. Software culture proposed a seemingly magical realm of possibilities. A single line of code and a floppy disk could destroy/save the planet. In the retro version of this Build, #hackerman could travel through time and change history (Kung Fury, 2015).

Steps to reproduce:

Select Preferences -> Cultural History -> Speculative Imagination -> Critical Perspective

How to trigger the bug.

You can trigger the bug by raising questions about software development practices, documentation, inclusivity, and diversity. Is this software challenging to read? Does it allow contributions? Who is it leaving out? Is this algorithm increasing social integration, community, and relations? Does this code enable thought or dilutes it? How can your software promote reciprocal, balanced relationships with technology, nature, other living beings?

Actual results (what are the consequences of the bug)

Loss of a microcosm-macrocosm sight to balance the relationship between technology and social, political, economic, and environmental interaction. Developers from peripheralized geographies and industry spaces are marginalized in the technosphere.

Expected results (what should have happened without the bug)

Programming could become a common social practice. In it, an open witches' sabbath or a collective of shaman-distributed-coders could perform hallucinatory code exercises and explore new machine-human-nature interaction protocols.

Alphabetic normalization depletes trans-discursive possibilities by Femke Snelting

Summary (describe the problem in less than 60 characters)

Alphabetic normalization depletes trans-discursive possibilities.

Component (in which context does this problem exist?)

The Alphabet Issue affects all networked computational devices, infrastructures and platforms, especially those connected to and with the internet. Or as font engineer Behdad Esfahbod phrased it, “any 1 who reads on computer”.

Version (historical background of the bug)

The Alphabet might already have become an Issue when in late medieval times, Western Europe saw the emergence of an efficient solution for reproducing Latin languages: the printing press. The Gutenberg Galaxy assumes that language is made up of a limited set of autonomous, unchangeable elements; the letters of the alphabet. These atomic elements can be treated as a recombinable and reproducable set, ‘movable type’. A subtle consequence of introducing this pre-runner of agile computing, is that characters can be treated as separate from their rendered expressions, meaning that ultimately their semantic meaning is not anymore tied to a specific form or typographic rendering. Both principles had deep implications for computation as they form the basis of the ASCII table, a reduced alphabetic interface that also in contemporary devices regulates communication between hardware input and programmatic events. Decades later, The Unicode Consortium tasked itself to mitigate the limitations of ASCII both on an institutional and on a technical level. Unicode promises to “provide a unique number for every character, no matter what the platform, program, or language is” but effectively, and in addition to the assumptions above, still reduces language in the digital domain to a variation or extension of English. These processes of literally ordering the alphabet radiate out into national standardization bodies, Open Type specifications and Open Type Formats. Over the years, these bodies have found more and less successful solutions to patch the many issues with so-called non-Latin writing systems, but not at any moment are the colonial assumptions underlying the system of encoding being questioned.

OS (within which legal~societal~computational~material system is this problem)

The Alphabet Issue formats computational logic as a specific form of logo-centrism, it powers an extractivist understanding of language, it keeps colonial structurings in place and fuels capitalist hubris.

Description (larger detailed summary of the bug)

The Alphabet Issue operates on many levels. A non-exhaustive list:

  • With English as an exception, many writing systems use special combinations of letters and accents. Only with some effort can they fit into a single character based paradigm. As a result, most languages other than English struggle with the standard to some degree.
  • The basic organizational grid of the Unicode standard is completely alien to combinatory writing systems such as Arabic or Persian. It takes additional software solutions to make these writing systems work.
  • Computational devices assume and actively promote alphabetic language as the basis of all communication; in other words, computation is bound to a logo-centric drive. While the possibilities for life are increasingly impacted by what computers do, the bandwidth to communicate is narrow.
  • It binds device- and software-development to a particularly Euro-centric perspective on language. This cultural bias is convenient because when language is imagined as a limited set of independent, recombineable codepoints, it becomes extractable like any material resource for Natural Language Processing, Machine learning, Big Data etc.
  • In computational devices, the characters in the alphabet are understood as in essence detached from their letterform. The Unicode codepoint U+0041 can be rendered as A or A or A and actually also as A. This u) dematerializes language because it assumes that language and its representation can be separated and e) ignores the interdependencies between letterforms that are at the basis of many writing systems.
  • This also means computing as it stands is badly situated for coming to terms with its own implications and worldings because de-situating and anti-materialism are at its core. When the relation between character and glyph, or worlding and language are structurally disentangled, digital communication is ultimately a techno-colonialist force.
  • The Alphabet assumes a normalized order (a, b, c) that is almost without exception figured as-easy-as-abc. This empowering simplicity is reiterated in many alphabet songs such as the Jackson 5 lyrics “A B C / It's easy as, 1 2 3 / As simple as, do re mi / A B C, 1 2 3 / Baby, you and me girl”. Or think of the way Frrrench philosopher Gilles Deleuze omits to question the arbitrary rigid order of the alphabet that organizes the interview which Claire Parnet titled L’Abécédaire de Gilles Deleuze. It also structures trans*feminist dictionary works like Bug Report: Tuning to trans*feminist System.Crash, to which this text is a contribution.
  • Trans-discursive possibilities are limited since any interspecies conversation needs to stay out of computation or accept at best a translation from and to alphabetic norms.
  • The internalization of these norms is intensified by the promise of their all-encompassing completeness and universal openness. It is this universalistic arrogance posing as innocence that spawned the outrageous renaming of the Google’s subsidiaries into The Alphabet Companies.
  • "build ID" (which non/humans does this effect most)

    Humans speaking and writing in other languages and scripts than English. Additional Builds & Platforms (which other contexts are effected? what other positions does this have implications for?) Non/humans communicating in other ways than with the help of alphabetic language.

    Steps to reproduce: how to trigger the bug

    The issue with this particular bug is that in the current state of computational ubiquity, it is always already triggered. The consequences of The Alphabet Issue are therefore as invasive as they are invisibilised. Did I already bring up how many instances of the Alphabet Issue are operating around the filing of this bugreport? The Alphabet hardly ever raises an Issue and is extremely difficult to address. It covers up and is covered up by deeply colonial, Euro-centric and damaging assumptions about communication, computation and culture. actual results (what are the consequences of the bug)

    See above.

    expected results (what should have happened without the bug)

    It would still matter how letters arranged, but not everywhere and not always in the same way. We would invent numerous ligatures and diacritics without end; form and codepoint would amalgamate instead of s e p a r a t e , rendering trans*feminist orientation tangible. With effort, focus, and energy, spelling shifts from ortography into casting spells, now making arrangements that definitely definitly definitely defiantly would change what can be circulated by and between interdependent entities.

    P.S. This bugreport is based on research with Peggy Pierrot and Roel Roscam Abbing for Modifying the universal, and also OSP’s Dingbat Liberation Fests. The Alphabet Issue was sparked again by the recent emergence of critiques on the coloniality of font systems and digital typography as documented by Pierre Huyghebaert.

    bug in my dictation app by Pinar Tuzcu

    Summary:

    My keyboard dictation app does not recognize so many words when I am talking to it. It either crashes or puts punctuations like ‘!’ or ‘?’ As I am not able to write without dictation, this is frustrating. The error occurs especially when I say ‘heteropatriarchy’ ‘gender equilibrium instead equality’ ‘colonial destruction on ecology of gender in colonized societies’.

    Component:

    General

    Version:

    This bug existed in early Version of Mac OS such as 10.11: "El Capitan" and now I have Mac OS Catalina version 10.15, it still exists. I hoped for some improvement, but I see that despite of the fast developments in technology, such bugs remain integrated in the system.

    OS:

    All Operating Systems

    Overview:

    My keyboard dictation app does not recognize many expressions that I am voicing. It either crashes or puts punctuations such as ‘!’ or ‘?’ instead. As I am not able to write without dictation this is frustrating. The error particularly occurs when I use the words or expressions such as ‘heteropatriarchy’ ‘gender equilibrium instead gender equality’ or ‘colonial destruction on ecology of gender in colonized societies’. When I dictate, for instance, “colonialism did not improve the colonized societies in terms of gender equality, this claim is just a white lie” the app puts a question mark in the sentence, arbitrarily. Or when I dictate “in fact, European colonizers irrecoverably damaged the ecology of gender, that is the equilibrium existed among diverse sexes in these communities in pre-colonial times” then the app inserts exclamation marks in the sentence. Once I said, “a Nigerian anti-colonial feminist Funmilayo Ransome-Kunti wrote in her book For Women and The Nation of Nigeria that ‘we had equality till Britain came’” and the app collapsed completely.

    Build Id:

    This bug affects my bodily integrity because I need the dictation app to be able to ‘write’. On the other hand, I feel like the system does not recognize what I am saying. Or maybe, as I am not a native English speaker, it is the matter of how I am saying it. I don’t know. But there is definitely a recognition problem. Yes, what kind of trouble in the recognition system might cause such problem? A system that executes my ‘sayings’ not understandable.

    Additional Builds and Platforms:

    The vocabularies, speeches and accents that do not fit into the heteronormative patriarchal syntax and Eurocentric prescripts are considered to be unrecognizable, marked as unsayable ‘?!’. Such a bug erroneously decides what is intelligible what is not; what counts as ‘appropriately expressed’ and what does not.

    Steps to Reproduce:

    I turned on the app and start to dictate: “The European colonizers ‘empowered’ only a small group of people in these communities, those whom they identified as ‘man’ through their hetero-patriarchal white gaze and according to their European binary standards of gender. They armed those whom they took as ‘men’ with weapons in order to build military troops for both integrating the colonial orders in these societies with violence and defending their colonial settlements against other countries. This ignited violence within the communities since the armed group of ‘men’ felt powerful over the unarmed and other genders. Moreover, the colonizers redistributed the land after stealing it from its real owners to the local and indigenous ‘men’ by ignoring the fact that the land in these communities was hand down from generations to generation according to the maternal-kinship-line. After destroying the ecology of gender, the equilibrium existing among different sexes, in these communities, the colonial settlers forced the local and indigenous population to adapt, what they called, ‘gender equality.’

    Actual Results:

    The application crashed.

    Expected Results:

    Voice recognition apps such as dictation apps should be able to detect the alternative ways of telling stories that do not fit into the heteropatriarchal neocolonial pretext. It should be attentive to the codes, to the words, and to the social worlds that did not make it into language; to the accented expressions that did not pass from one language into another. The app should make our wor(l)ds survive, if not, then let it collapse.