Patternflow

---
title: "Stopped"
date: "2026-05-23T07:00:00+09:00"
excerpt: "Seeing it as unfamiliar"
draft: false
---

This week feels like I did not do much. To promote Patternflow, I posted on Reddit, made patterns every day, uploaded them to Instagram, and organized them on Discord too. For optimization, I restructured the firmware code and added OSC communication. I also recalculated costs while preparing for sales. Writing it out like this, it is not that I did nothing. And still, I keep feeling like something has stopped.

I went back into Crowd Supply. I had heard it would be hard to get in, so my expectations were low, and after applying I more or less forgot about it. But there was a reply the very next day. They asked why I was going with a kit instead of a finished product, and how manufacturing would be handled if the funding project moved forward. It felt like a pretty positive signal. So I replied right away, but three days have passed and I still have not heard anything back. That makes me anxious. I keep wondering if replying after four days was too late. I check the Crowd Supply site twice a day to see if a reply has come in.

I also made a proper quotation. I needed to stop using the AliExpress estimate and make something more realistic, but it was difficult and expensive. PCBA was the most important part, and among the components, the rotary encoder price went up dozens of times. So for now I am going to buy one expensive option and one relatively cheaper option in small quantities, test whether they fit and work on the Patternflow PCB, and if they do, place a PCBA order right away. In any case, experiments cost too much. The problem is that I do not have money.

Through Art Korea Lab business consulting, I am getting advice from Director Kim Kevin at P&G Ventures. Following his advice, I added labor cost to the production cost and calculated a worst case. It came out to 160,000 KRW. If I add SG&A and payment processing fees, selling at the originally planned $129 would lose money even before international shipping. So I am worried. But there seems to be room to reduce it. PCBWay has also agreed to provide sponsorship, so I need to handle that conversation well, and the most expensive part, 3D printing, can be reduced if I use the school makerspace instead of outsourcing.

Still, I do not think things will move as quickly as I hoped. I wanted to send Patternflow to at least around 20 of the 70 people who said they would buy it within a month, but that seems difficult. I heard that receiving the PCBA order takes about 25 days. So I think I need to test once, verify it, and only then move forward. During that waiting time, I plan to try package design. I had thought about asking another senior designer before, but I cannot ask for it for free, and I do not have money, so I will probably try doing it myself for now.

There is good news too. Patternflow has its first collaborator. Zhuoran. She is interested in pattern generation, especially image-to-pattern conversion. Looking at the texture of her work and the tools she uses, I thought she fit Patternflow really well. I am grateful that someone like that reached out first. I was the one who suggested collaborating, but honestly, it was not easy. Saying open source, open source, and sharing something I made was easy. I learned by looking at and using other people's work too, so it felt natural. But a collaborator is completely different, is it not?

It felt a little like the feeling of a parent sending a carefully raised child to daycare. I was very anxious, and all kinds of worries came up. I did not have anyone to ask about it, so I asked AI. "Can we recover if everything breaks?" "How should we split revenue?" Things like that, useless questions. After worrying as much as I wanted for about a day, I felt okay. I decided to move at my own tempo again, and it felt good, like Patternflow had become a real open-source project. Two days ago, as part of an international shipping test, I sent Zhuoran one of the Patternflow OG units. It was the only clean one among the two I had. So if you look at the content I post on Instagram now, I am filming again with the first Patternflow, the one missing a few LEDs around the corner. Shipping was incredibly expensive. I thought it would be around 30,000 KRW, but maybe because I wrapped it in ten layers of bubble wrap out of worry, it came out to 60,000 KRW.

I am also getting another kind of help. Professor Elon Choi agreed to help with 3D printing. Until now, because the P1S size was not enough, I had to print Patternflow in separated parts. But he said it might be possible to print it in one piece, and that he would reverse engineer it himself to make integrated printing possible, while also improving additional details. He is an expert in that field, so I trust him. We also agreed to run a workshop later. I cannot even pay him properly, so I am grateful that he is helping.

And now the things I need to do are clear. People are slowly starting to make Patternflow. Maybe many more will suddenly appear. I need to gather them. It may be okay to let them scatter, but for the Patternflow open-source ecosystem, they are incredibly valuable people. I need to make a place where they can make patterns, share them, modify them to their own taste, and share those too. The GitHub page also needs to guide contributions more clearly, and the place for communication needs to be cared for better. I need to become a gardener now, not just a developer.

To do that, I need to look at it as something unfamiliar. Right now I know too much about Patternflow. There are too many things that feel obvious to me, and I need to throw those away. I need to look at it as if I were seeing it for the first time, imagining how I would feel when encountering someone else's open-source project. Honestly, I have never contributed to other open-source projects. I usually found it hard to understand what I was supposed to do, so I only used them. I think maybe that is because there is not enough visual guidance. I should try to fill those gaps. After all, I am a visual designer too. It sounds fun.

At the same time, I need to fix all the documents I had not been able to care about. While recalculating the BOM, I found one critical issue. The 2×8 connector for the HUB75 was actually vertical, but it had been listed as horizontal. So anyone who bought parts by looking only at the BOM list instead of the image would have to order again. I am very sorry. I paid a little too little attention. I need to fill in all these small but fatal user experience gaps. I need to refine the details. I also need to change the overall structure. It is hard. Well, let's try.