Patternflow

---
title: "headache"
date: "2026-05-14T01:30:00+09:00"
excerpt: "What did I do this morning?"
draft: false
---

I'm losing it. My head is too crowded — I've gotten to the point where I can't remember what I did this morning. It's gotten frighteningly bad. *"The feeling of slowly disappearing,"* I guess I could call it.

I had my Omnifit stress test today at counseling, three weeks after the last one. Brain activity is still in overload. And my stress index, which used to be in the normal range, is now in the *very high* zone — both readings hit the end of the bar graph. I'm overworking my brain. There are too many thoughts. I know I should rest a little, but I don't know if I can. I think I was in a similar state two weeks ago. I said I'd rest, and I didn't. This time will probably be the same. But I don't know. I really don't know. Is this right. Or wrong.

---

## Patternflow is a big opportunity

I think Patternflow is a real opportunity. Making my own product, having people recognize it, earning money from it — that's been a dream of mine for a while. And the potential seems much bigger than the moon-jar frames I sold briefly and then gave up on. The person who properly went to market with those moon-jar frames made over 200 million won. And Patternflow has gotten a bigger early response than the moon jars. A completely different scale.

So if I do this properly, couldn't I produce something extraordinary? It feels like the kind of opportunity that could solidly fund the next round of research, art, and study. So I'm pouring all my effort and passion into it. I was already an impatient person; now I'm even more impatient.

I've met with a few business experts lately. We talked about Patternflow, and all of them spoke positively. But their timelines were too long for me. By "they" I mean experts in the business field, so I think it's fair to say they represent the existing system. Securing and defining the early customer base — minimum two months. Real sales they projected for the second half of this year, or early next year. Using data to gather funds through government support programs, then actually starting the business next year.

Too slow. In my plan I want to be selling within a month. Whether through crowdfunding or selling small batches — there are already lots of people who said *"I'd buy this"* in surveys. I hate making them wait. I want to make it fast and get it to them fast.

---

## I've stopped leaning on people

My mindset has shifted recently. Before, I wanted to get help from experts and other people. So I'd ask with expectations, and the moment I got no response, or felt rejected, I'd be deflated and lose motivation. That caused delays.

Now I think of it as fundamentally something I do alone. No matter how much experts attach themselves to help, the actual doing has to be by me anyway. And that's right. **For something I made, no one is more of an expert than I am.** I should ask them for opinions, not beg for help. That's a little hard. But I like it. If it works out, my credit. If it doesn't, my fault. Well, actually, mostly luck.

---

## The question — stop everything and throw myself in?

It's hard. It's not just Patternflow — I'm also taking school classes, and I'm the development team lead for the exhibition club on the side. We're still in the planning phase so the time demands aren't huge yet, but it's coming. I don't know if I'll make it through that. Should I just explain the situation, quit, and focus on Patternflow?

In counseling today, they said of course that's possible. No problem at all. And honestly — if someone else in my position said this big chance came up in their life, I think I'd cheer them on and let them go. Even if one person leaves, things still run. That's what community means. So why is it hard to do for myself? Am I feeling some special sense of responsibility, or anticipating guilt, or do I actually not think Patternflow is *that* big a deal?

When I keep going down this path, the same applies to school. If this is an opportunity of this scale, if I want to do it properly, shouldn't I stop everything and pour myself into it completely? Shouldn't I? But what if I fail then? Then I've ruined everything. No — if I pour everything in I can make sure it doesn't fail. Really?

Ah no — there's a separate reason classes can't be subject to this kind of judgment. **They're fun.** I love every class, and I love the people in them. Then what about the exhibition club? Now everything becomes a contradiction. I don't know. My head won't turn over.

By the way, it's the school festival period right now. It runs three days, yesterday through tomorrow, and it's the last festival before the construction starts, so I was planning to enjoy it all. But today seems hard. No joke, I might collapse if I run around, so I'm just going to spit out some writing and sleep. I have to hang the laundry I started, so I need to write for about two hours. Why did I start the laundry. I regret it. I checked the time. One hour. Nice. Okay, but let me at least scribble a quick journal entry before sleeping.

---

## Pulling patterns out of AI, and why it's no good

Enough digression — back to Patternflow. Lately I've been *pulling* patterns out using AI. *"Pulling out"* fits better than *"making."* I'm literally running the prompt through three different AIs at the same time, plugging everything in, and picking the least-bad result. I don't know if this is a good method. It's not a desirable one. If I were at least comparing the results across prompts and studying it, fine — but I'm doing nothing of the sort. No improvement, total reliance on luck.

That said, I'm not saying AI-based pattern generation itself is bad. I see it as a forced choice. It's not that I don't know the joy of studying formulas and techniques and finding novelty through creation. But I don't have time for that right now. Every single day I have to make content, show it to the public, show it to the people who might contribute to Patternflow. Fast, plentiful, consistent. To be consistent, it has to be easy. If I went through the agony of creation every day, writing code by hand, I'd be flat on my back within a week. Guaranteed.

---

## The word "business," and the cafeteria

I can't focus only on patterns either. The current stage is less research and art than it is business. There's so much to take care of. I need to constantly manage channels and write. Writing includes this blog, and also business plans, which I genuinely don't understand the point of and find tedious. Why on earth do I have to write those? I don't get it. But I still have to. Because I need funds.

I don't even have money to experiment with right now. I'm getting by on the cheap school cafeteria meals every day. If I made a pile of money, I'd want to eat expensive healthy things like salmon salad every day.

I've sent workshop proposals to a few places. I've always liked teaching. And if I can earn some money lightly while doing it, isn't that the best? So I went to a meeting recently. If it goes through it would be amazing. Anyway — these kinds of personal activities also have to keep going. A solo exhibition too. Right. My goal this year was to hold a solo exhibition of my own work. At this rate I think I can.

I'm starting to run out of things to write. The washing machine won't stop, so I have to keep writing too. I want to do crowdfunding. I want to make some absurd outcome happen with global crowdfunding. Ah but I shouldn't try that on the first one. Well, let's do several. Crowd Supply once, Kickstarter once. I hope it can go that way.

---

## Back to AI pattern generation — actually, this matters

Oh right, I need to talk more about AI pattern generation. I'd completely forgotten. My head really does seem broken.

There's one important reason I have to research the technique of making patterns with AI. **I want the people who buy Patternflow to make their own patterns.** But suddenly telling them they have to write code themselves, learn this language, learn that algorithm — that doesn't make sense. It has to be made easy. Thanks to AI, it's definitely possible. Wait — is it? That depends on what I do. I have to keep experimenting with what each prompt produces, how to combine them so they come out properly. Abstract it, find the patterns within the patterns.

What do they call this? Ontology? It feels like that's what I'm building. I'm not sure if that's the right word. I picked it up somewhere and I'm just using it. If I'm wrong, I'm wrong.

Hm, I was going to write something but I've lost it. I have to wash up before sleeping anyway, so maybe I should wash up now and write the journal entry too. Hang the laundry and collapse into sleep.

---

## Meta — Claude refused

Lately I think the blog has been feeling messy and lazy. Sorry. But I can't help it. The state I'm in right now isn't a state where I can write properly. I'm just spitting out whatever comes to mind. Ah, but I'll have AI clean up the structure and expression a little. Well, I've told it to keep my tone, so probably nothing big will change… It'll still be hard to read. You'll be thinking *"what is this person on about."* It's just that kind of post.

Oh, the Claude weekly limit's been hitting really fast lately, so using AI has gotten a little scary. Let me earn money quickly and switch to Max.

---

So — I asked Claude to clean up the post and translate it, and it refused. It said this post feels too different from the previous ones, and that even I'm aware of it, and told me to look at it again when I wake up tomorrow. But I don't want to. Why not just upload all of it? **Isn't this also part of making something?** AI is the generic option, so to do something unique maybe I should go against it. That sounds more fun. I'm going to add all of this to the prompt and ask again.

Whether I'll regret it or not, I don't know. So I'll just try it. That's how I live. **You just jump in.**

---