Micro Challenge
Course Details
Name: Micro Challenge I & II
Dates: 04 February to 07 February 2025 & 04 March to 07 March 2025
Faculty: Santiago Fuentemilla Garriga & many more
Micro Challenge I
For the first Micro Challenge, I worked with Auxence Daillen and Ramon Prat, we spent the first day trying to tie our areas of research together into a project. Eventually we settled on the idea of trying to continue the research I had done for my first intervention this term. We wanted to try to spam particular YouTube videos on a local WiFi network to influence the suggestion algorithm.
We decided to create a narrative around this 'digital poison' and disguise the device in a custom bottle that could be brought to a cafe or other public place with a WiFi network to 'infect' the network.
Reflection
We encountered some challenges in this project, the first being that we struggled to choose a project related all of our interests. Unfortunately, I found this very frustrating and and struggled to get past the feeling that we had 'wasted' the first day trying to decide what to do. The project we eventually chose was something related to an investigation I was interested in from a previous intervention, though, so I was excited to get to explore it further.
The work itself that we did was interesting and we encountered many challenges throughout the process. We tried different processes related to which hardware we would use, either the Barduino or the Raspberry Pi. I struggled with the sense that we were redoing work, pivoting to a new system after already putting a full days work into one system. However, with some time since that frustration to reflect, I can recognize that this type of pivoting is all part of the process.
Additionally, a challenge we encountered was that the project did not work in a few fundamental ways that made it difficult for me to feel satisfied about what we had achieved. I had a vision in my head about the scope of our project. My intention was to have a device that we could connect to a local wifi network and 'watch' a series of specific videos over the course of many days. Then we would be able to monitor the 'infection' of the network by seeing what content was suggested by the algorithms on our own devices and accounts. This theory was based on some experiences I have had of videos that a friend showed me on their device showing up on my device a few days later. However, during our prototyping phase, we were told that YouTube in particular has certain encryptions in place that would almost certainly make this theory not actually play out. I was determined to keep trying though, because even if it didn't work, that would be interesting insight as well.
I started digging into the YouTube API to try to figure out how we could monitor suggested videos, however quickly realized that there are, in fact privacy protections in palace that limit what someone can see about their own account even. So, I transitioned over to an investigation about how to log information about the videos we were watching using the Google Sheet API. While we were unable to get any of the functionality we envisioned in working order, I did learn a bit about using APIs and was eventually able to find value in the project overall, despite it not being anything like what I had expected to be doing or being very related to I had wanted to investigate.
Another point of frustration with this challenge was when our documentation on Hackster.io was deleted somehow by their servers. I have learned, now to have backups of the documentation in another place, perhaps on GitHub, so that work does not need to be repeated when things go wrong in ways beyond our control.
Overall, this Micro Challenge was perhaps more mentally and emotionally challenging than technically challenging. While we encountered many technical challenges, working through my own frustrations was certainly the biggest challenge I had to tackle. However, looking back at it, I realize that I was able to learn useful skills, just not necessarily the ones I expected to learn.
Micro Challenge II
For the second Micro Challenge, I worked with Carlos Silveira and Flavio Grimaldi.
Reflection
For Micro Challenge II, I was determined to not spend an entire day trying to pick a project, so I joined a group who already had a clear idea of what they wanted to do, and set out to bring skills I wanted to learn into the project. I was interested in learning about communication protocols between devices and ideally to integrate MIDI and audio control into the project. While I barely scratched the surface of the communication protocols, I got a good first taste of the challenges that can arise with these kinds of systems.
Additionally, through this project, I found myself more and more convinced that ChatGPT is only marginally helpful when it comes to writing code. I have always found it frustrating to attempt to understand someone else's code, especially when it is not documented very well, and I found this a challenge when trying to integrate the code my group members were writing into the code I had written. Because there was some dependence on generated code, it was challenging at times to follow the logic and to figure out how to integrate these codes together. However, at one point, I used ChatGPT, despite my protestations, to help me figure out the logic of a code piece I was struggling to write. So, overall, I can recognize that these tools have a place in coding workflows, if they are used carefully, consciously, and with context for what the resulting code is doing.
This Micro Challenge still had significant moments of frustration when things did not come together as we had hoped. As usually, given the short time frame, we had to scale our ideas back significantly, and then scale them back again when time started to get tight. As with many of these sorts of projects, by the end it felt like if we'd had 'just one or two more days' we could have gotten it working better, however, part of the value of these short projects is to learn to be okay with things that are not complete, not meeting the vision. This is something I still struggle with.