A Study of Light
A Study of Light is a System to automate production of art content revolving around the Magic Milk Experiment physical phenomena. Automation is a major theme. The extrapolated load on myself must be minimised if the content generation is to be high-volume and sustainable. This is a very-full-stack project. The different areas are outlined in the following pages, from low- to high- level:
- aSoL Experiment Phenomena
- aSoL Mechanical
- aSoL Electronics
- aSoL Controller
- aSoL Communications
- aSoL Backend Systems
- aSoL User Interfaces
- aSoL Cloud Architecture
- aSoL Content System
# The Idea
The idea for this project came to me somewhat accidentally 22Q1. I had bought a science kit to do with my sister, containing 3 different dyes. Some time later after having done everything in the kit, I came across the Magic Milk Experiment online while searching for more things to do. We used the petri dish and dye from the kit, and some milk and dish soap. I was very surprised at the elaborate phenomena observed. The level of detail, beauty, and randomness in the results were, and still are captivating.
Some weeks passed. After talking about art and NFTs with a friend I got thinking about how I could get involved in such things. Eventually this thought connected with the previous experiment and I had the idea of creating a magic milk NFT collection, around 100 different ones that could be listed on an NFT site. So I bought a ring light, found a nice black bowl, and got set up. It sounds silly, but I remember being awe-struck when I first saw that brightly lit white circle from the top-down view; a blank canvas of infinite potential, which would so thoroughly expand my creativity and technical skills. The first session was a pain with lots of manual overhead, but it got me thinking hard about what would come next.
I quickly realised a couple of things:
- I did not have the discipline to repeat that 100 times.
- I could probably build a system to do it for me, so I don’t have to.
Before building anything, I had envisioned a robot that could be remotely controlled by anyone on earth, for a fee, to generate their own art and NFTs on a blockchain, with a fully automated content production system to flood the internet with high-quality content, capturing public attention.
This was a bold, somewhat overwhelming vision. Given my past history of project-hopping and “lack of discipline”, I was half expecting that I would get bored and give up after a while. I was also concerned with the risk of getting lost in the technical aspect, building an Over-Optimised & Under-Utilised system that I would never have the courage to actually use. So I planned to continue with manual sessions and make the engineering subordinate to the art.
The manual session idea didn’t last, but after 14 months of work I can say I’ve solved my project inconsistency issues and am on track for a scalable end-to-end system by 23Q4.