Queering our keyboard workshop
first draft
(most recent version is at the bottom of this page)The workshop will investigate how your computer knows which character to display when you press one of the keys on your keyboard. By making our own letters and making a keyboard out of them, we will encounter some of the technologies, standards and protocols involved in this process, and we will touch on some of the politics surrounding them. Our self-made keyboard will show why this part of our computer, phone, or other device, is maybe less self-evident and fixed then we might realize.
Planning:
- Open the day with a quick round of introduction. And a small discussion on how we think our computer interprets language when we type.
- give an overview of what we'll do that day and an introduction into the topic. why closer examining our keyboard is a queering act.
- making our own 'script' / 'language' by cutting out shapes from erasers. This is to mimic the process of creating type casts. An intuitive introduction to the process of industrial typesetting of the latin script. From there we can make the connection to how that process formed the basis of digital language. The 'script' we're making doesn't have to resemble anything like a working / existing language. It might actually hit the point home even more that to the computer, it does not matter if the shapes we make are meaningful or interpretable as readable language.
Depending on how many we are, everyone can create 1 or 2 'letters' that we can bring together to make a collective 'script'. And we can use the erasers as stamps to show the basic logic of latin-based industrial typesetting.
- we can digitize the font by scanning it as jpeg, either put it in fontforge directly or use another program to make vector shapes out of them.
- lunch! break for some food, (for me it would be interesting to see if discussions around the topic form naturally)
- short overview of what we did so far and how we will continue the afternoon.
- make our letters into a typeface using font forge. I'll find a protocol to share all vecotrised letters between everyone. everyone makes their own version of the font file in fontforge. There's a few ways to go about this part that depend a bit on what is possible on the technical side. So this might change a bit, but these are some options:
1. we stick to the latin character set. In the place of the unicode codepoints for latin, we put the letterforms of our own script.
We generate the font into a .otf file. In keyboard software (important note on this at the bottom), we add the font to the existing Latin keyboard and export the keyboard.
-> this should be the most straight-forward way to do this. The encoding is done in the fontfile rather than the keyboard itself. But it should be able to type with the font we assigned to it.
2. we use Unicode's Private Use Area code-points.
In font forge, we add PUA code-points to the fontfile, and place our characters on those. in the .otf file, you'll be able to write in both latin characters and our made up script. In the keyboard software, instead of replacing the latin charaters, we assign PUA code-points to the keyboard keys. We add our font to the keyboard software, and add the PUA code-points to keys when Alt is pressed. Then, add our font, and tell it to interpret the empty PUA values with the characters we assigned to them in the font file.
-> haven't managed to make this work so far, even though documentation says this should be the way to do it. It would mean that typing with the keyboard in latin would show latin characters that can be interpreted by the computer as such, writing in our script would show the characters when typing while ALT is pressed, but the computer will only recognize the empty unicode values.
--> Probably need to assign the right language to the keyboard. The language that contains the unicode values of the PUA?
3. we do the same, but instead of PUA code-points, we use the standard latin capital letter code-points. The keyboard would be able to write in latin lowercase and our script show our characters while shift is pressed, and the computer will interpret it as capital letters.
In any of these cases, throughout the process we will have to deal with the unicode standard, the ISO standard for language names, the .otf file format, how the logic of industrial typesetting ( the erasers ) translates digitally, and why that means that Latin will necessarily always be the default script system that other writing systems or practices must adapt to in order to work digitally, and what organizations are in charge of regulating this through these standards.
I'm hoping that this visualizes at least that how we use language digitally is not a given and that the politics surrounding it are quite significant. There is actually a thin line between fiction and reality when it comes to language, and these systems at the intersection between technology, politics and aesthetics decide for a big part how we come to imagine some use of language as fact, formal, fixed and others as made-up, backward, or fake.
- the last step would be to create a collective poem using the keyboard with our made-up script. This will probably be a bit improvised, since it will depend on how far we all get with creating the keyboard. Also need to make a protocol for this.
- and then we end the day. If people like to stick around we have some drinks, have a bit of an after discussion if there's some energy left, or we play some Wii.
Timeline wise, this will probably take a full day. I'll work the schedule out it detail when I figure out the technical parts.
Setting:
We can sit around the big table in my studio. Extension chord in the middle of the table. Coffee machine and fridge close by! I'll bring erasers and stanley knives for the first part of the day. For the second part, participants will bring their own laptop.
I need to figure out the wifi, since we all use mobile data in the studio. Either people will have to use their own, or I'll give my hotspot.
Date:
21 / 22 march
equipment:
- extension chord
- scanner
- wifi / hotspot
- laptops (everyone brings)
- monitor / board to sketch
- water
- lunch
- drinks
crowd:
- How many ppl can I host? -> thinking 8/10 ppl max.
- How many ppl will be familiar with typography terminology / have experience working with code / terminal?
- studio mates
- friend groups
- if group size allows it, whatsapp / signal groups. Dear Fem Rotterdam (queer / fem resource sharing chat), Snackbar Frieda chat (squat / anarchist community chat).
Communication:
- make lil' announcement banner. title, time, date, location, small description, what to bring (laptop, OS, (own data?))
Documentation:
1. A collective poem, typed with the custum keyboard & script. If it will be possible to communicate across devices, we use a collaborative software that can recognise the font to write the poem together.
2. all the sketches, references, etc. I use, or others bring up, that help understand how the custom keyboard works.
NOTE ON SOFTWARE:
There's a few options for custom keyboard software:
The most widely used free / open source option, with widest OS support is Keyman. Which is significant since this software is created by SIL Language Technologies. (More of my research on this organization so far at the top of this page). The short version, it boils down to this organization is continuing the colonial project at present, at the far margins of digitally represented languages. I'm debating whether or not to go in-depth on this organization, and how I feel about using this software in the workshop, and if I do, how to do it.
They also maintain the ISO standard for language names. They do regular applications to Unicode for adding 'undocumented' languages to the standard, developed the Open Font License, and made a bunch of other tools for digital language representation.
I've done some first tests for the workshop in this software so far. I think it's important to show, but need to research more on other options.
So there's the ethical point of using this on the one side.
The practical point on the other. A lot of other programs are restricted to a particular OS, are paid, etc. If I ask everyone to bring a laptop, how will I manage this? To be figured out..
Manetta's feedback:
- stay on 1 device to keep it smaller
- use python (can I do that?)
- emphasize the queering part: character encoding as character miming, who to mime back?
- give examples: ByeByeBinary
- it can be way smaller exercises, it can be something small
and my thoughts:
- can I use a python script to do the same / a similar thing as what I am trying with Keyman?
> it would be the most DIY way to edit keyboard keys, and would get around the need to use Keyman.
> through the command line. How intuitive would this be for people that are not familiar with the command line at all?
> why the keyboard? to focus on a digital interface that for most people feels intuitive / self evident / fixed. to examine why it is not. compared to that, the command line is not an intuitive interface to many people.
- what's the queering part? Needs to be more explicit. Maybe the main focus of the workshop. The exercise should be what illustrates that point.
> a more direct reference to Butler. Computers always work in the logic of the Latin, no matter what language / script a computer is set in, a keyboard is set in, or language is written. Multi-script character encoding is saying to a non-latin letter: take the place of the 'a', act as if you're an 'a', and work within Latin logic. From there we add extra code to make you behave like your proper script does. Arabic as an example: The Alif will take the place of the A, and placed in the tiny block that is A, but Arabic letters change form depending on their position in a word, and is written right-to-left. Extra rules need to be added to create that behavior. But even when presented with an Arabic text, the computer doesn't see connected letter forms, it still sees separate blocks. Each letter one block, one string of 0's and 1's. And it might display the content for the user from right-to-left, but it still reads the information from left-to-right.
Every technological development in digital language representation is some form of adding rules to a writing system that can't be properly represented by default. But the rhetoric around it, is that developments makes language representation more equal. That is something to question. It's at the intersection of technology, aesthetics and politics. It creates a double reality, like when Butler starts their essay on Gender insubordination by saying 'How come I can both be a lesbian and have to act like one at the same time?' Since Latin was imagined as the 'default', the 'original' writing system for computers, every other writing system, visually represented writing practice becomes a 'cheap copy'. The promise of true representation, true equality is a falsehood because of this. It's not possible. But since it's the only option, there is no way around it, it still matters to try.
Marloes' feedback:
- define scope, narrow it down
- situated approach, keep it close to yourself
- show examples that illustrate your story
- keep participants in mind: how much can they digest in a day? Make space for their input! Let them explore the material themselves as well.
!!! Ask for feedback at the end? How did they experience the day? Did their perception change? Maybe in ways I did not expect? How can I ask for this in a fun way?
- for preparing, prioritizing: Go through the whole day step-by-step.
Second draft