Behind the Screens: Jobi
An interview with livecoder and data scientist Jo Kroese (Jobi).
Jo Kroese, better known as Jobi, is a livecoder and data scientist traversing the Dutch hardcore and algorave scenes, for which they use custom software to create a proudly joyful mess of hard experimental pop.
In this interview we’d like to talk about their practices and tools, as well as work in the community and the way they cope with the radical and drastic changes in their practice resulting from the corona crisis.
What is your first encounter with live coding and what are sources of inspiration?
I’m inspired by the people I spend time with.
Which platform do you use and why?
When I was younger, I used to come home from school and immediately start composing at my computer. I was interested in complex rhythms from African and Indian traditions and wondered why rhythm was so absent from Western music, from baroque to techno. (Side note: I’ve come to believe that some of it is a colonial hang up: for some reason harmony and melody were seen as intellectual and rhythm was seen as fundamental and therefore base.)
Tidal Cycles allows me to make these complex rhythms so easily. Combining complex rhythms with the sounds and trademarks of pop and dance is still so fascinating to me and Tidal Cycles is the best tool for me for doing that at the moment. I also use SuperCollider to make some of my more niche sounds such as the distorted Gabba kick I use relentlessly.
How has live coding influenced your way of making things?
I’m naturally a perfectionist. Live coding brings a DIY, punk attitude of unmitigated, uncleaned expression of self. This takes me away from obsessing over how quantised a certain snare should be and how much eq (equalization, red.) to put on these vocals.
Live coding is making something in the moment to express something and not worrying about cleaning up that expression.
Did the isolation force you to make adjustments to your current practice, and can you elaborate on how these changes impact your work?
Making club music at the moment feels strange to me. It emphasises how far away we are from that sense of community and belonging, which brings me to make the music I do.
Instead, I have found myself focusing on my data science work in social and environmental policy. I am currently redesigning my website for the Radical Data Project, a place I explore alternative uses of data science in art and activism.
I have also become very interested in knowledge collection tools, which is why I created and released a wiki/personal notebook crossover I call tiddlyroam. Other projects include:
The Gratitude Machine
The Gratitude Machine is an AI learning how to be thankful.
I’m interested in how we can reappropriate technology to be kind, joyous and just. The Gratitude Machine project turns an AI called GPT-2 that was declared to be ‘too dangerous to release’ into a force for contemplation, ecological thought, communion and prayer. People can write ‘thank yous’ to it, helping the Gratitude Machine to learn how to say it’s own ‘thank yous’.
The picture included was the first text written by the Gratitude Machine. The first time I saw it, I had tears in my eyes.
I have sent extracts of the Gratitude Machine’s ‘thank-yous’ to friends and family. Some of them regularly look at it to help them.
It currently works locally, but I plan on turning it into an interactive website and also a portable light installation.
Mechanical Turk Project
I’ve also been interested in technological infrastructure that is made to obscure the underlying human relation. For me, the archetype of this is Mechanical Turk: you put in a question as if to a computer, a worker completes it as if a computer. From the side of the requester, the worker is left as only a user-id.
This got me interested in exploring how to imbue that relation with as much humanity as possible. Part of that is thinking about the task I set them: I’m currently paying people on Mechanical Turk to write poems about what they are thankful for. The other part is about how to make the terms as much in their favour as possible: I estimate I paid more than 10 times the average people pay for a task of my size and I also pay regardless of what they submit.
The power relations currently seem unavoidably in favour of the person paying. I’m exploring whether it is possible to get around that.
Is it possible for two people to connect with humanity through a system doing everything possible to treat you as two objects? Is it possible to bring some joy to another’s life through such an anonymous system?
Find more of Jobi’s work here:
— — —
This article is part of the 10 Minute Livecoding Challenge by Creative Coding Utrecht and Netherlands Coding Live — a series of events where digital artists and live coders create a piece in ten minutes.
The 10 Minute Live Coding Challenge is sponsored by Stimuleringsfonds Creatieve Industrie.