Faster, Faster

As of today, I've hit 150 subscribers, the threshold to move from pre-launch to launch on Crowd Supply. Ten over that, at 160. The sudden surge came from a pattern I made recently going viral, and the firepower is stronger than it's ever been. At the current pace, it looks like it'll land right around 300,000 views and stop. The pattern right before this one is a personal favorite and performed decently too. It's the same logic with only the base pattern altered, and honestly I'd worried it might be a little gross. My worry proved pointless; it's putting up the best numbers of all. It really is hard to predict what people will like.

This week I built a community site for sharing patterns like these. The list of things to improve and features to add is piled up like a mountain, but the usability itself is far better than sharing on Discord. The feature I think has the most potential is the fork. You copy a base pattern, change the code or the colors, and post it back up. The code side is honestly hard to get into, but colors are at a level an elementary schooler could manage, and it's fun. So it looks easy to build events around it and draw users in. The community site is up provisionally at patternflow.work/community. I'll keep developing it.

I want to start expanding the pattern-making features too. Supporting various resolutions and just laying the groundwork for adding layering to patterns would be enough. If I show that possibility, maybe people will build it themselves? That's the hope, but that isn't going to happen, and in the end I'll have to do it. After that, I want to make prettier, cuter patterns. The patterns getting good responses right now are, in the end, the cool, dark, what would you call it, psychedelic ones. The kind you'd see in a dark basement club. I want to make the opposite, peaceful and quiet ones. Put characters in… what I'm saying, in the end, is that I want to do pixel art.

Sound is a real headache. I haven't run many experiments, but based on my experience so far, the ceiling on the auditory side is very high. When graphics and sound, the visual and the audible, come together well, the response is excellent. It's just hard. There's no way but to keep making things and grow the instinct for it. The theoretical route I don't want to do and find hard, so nothing to be done. In that vein, I'm giving up on studying with the Designing Sound book for now. I'll come back to it when I can actually imagine certain feelings. Right now I can't even imagine what I want to make, so in the end no interest comes. It isn't fun.

Back to Crowd Supply: now I'll be preparing for the launch. I have to get real quotes, set the price, and rewrite the project page. Honestly I don't want to do that stuff. It's such a pain, and I wish Crowd Supply would do it for me. I just want to make patterns and do the fun parts. But I have to. The USB-C adapter needs revising too. According to an expert with plenty of years behind them, the current USB-C adapter is at a difficulty level where hand-soldering is impossible. So it wasn't that I lacked the skill! A relief, and a bit of a letdown. Actually, from the standpoint of writing the build guide, it's good that it's now clear. Build it yourself with the 2-pin screw terminal; USB-C only via PCBA. I can split it that way. Still, the revision is needed. Because I can't tell whether the shorting and burning is a soldering fault or a structural defect. There's a real chance it's structural, which is the problem. I'll fix it, what else can I do. Someone already proposed a fix, so I just have to try it. The details are here: GitHub Issue #221.

I think I've written enough about the current state and the plans ahead, so let me look back a little too. I did pass a milestone with the pre-launch succeeding, and there's a lot I felt in it. Naturally, I was impatient. More than five times a day I opened Crowd Supply's project management page to check the subscriber count. I memorized the number every time, and when it was the same on the next check, disappointment and anxiety would set in. I was anxious it would stop dead there and become a failed project. It was an absurd worry. I pre-launched on June 28th and passed 150 on July 24th, so it was under a month. That's incredibly fast. But in the middle of it I couldn't feel that speed at all, and it only seemed far too slow.

To promote it I wrote a great many posts. Counting the Arduino subreddit, where I was removed for being promotional, I posted to about five subreddits. I wrote the project up for Hackaday, and joined CAN, which I'd only ever admired from afar, to ask them to run a piece. I redid the PCBway community project with the new version, and shared the project on MakerWorld. I also edited and uploaded a video introducing Patternflow. At first I expected high view counts, treating it as a YouTube piece, but no chance; there was no traffic at all, and the version I put on Instagram with just subtitles added actually did better.

In terms of raw firepower, Instagram pattern content is better. But writing the project up and sharing it with materials had its upsides. Editors at big communities take my writing and materials and use them. There's no reaching out first to ask permission, none of that. They just take it and use it. Honestly I don't love it, but it's still a good thing. It gets the word out either way. And I take those pieces and use them as evidence proving the project's worth. One thing I learned here: organizing things so others can easily take and use them is a skill in itself. If a moment like this comes again, I think I'd analyze the writing of the people publishing on the platforms where my project's audience lives, and the material formats they like to use, and introduce it to fit that. It feels a bit complicated… honestly, I think anything works if there's sincerity in it. Even so, adding a strategic layer to make it faster and surer would obviously help.

If it keeps up this pace, maybe I can push the launch through fast and wrap it in two months? Then Patternflow goes out to the world around October, and whether I can manage the incoming people and build the ecosystem at that point will decide the whole thing. The network and community I have now, in the language of the field, is an atomic network, and it feels decent enough. It isn't as lively as I'd hoped, but that's natural since I haven't put in that much effort to begin with, and that makes me grateful to the ones who are actively there. I think I'll probably offer someone a moderator or collaborator role soon. The moment there's a good reason, I'll do it.

If I build the community foundation well and it becomes an ecosystem with steady interaction, I think that'd be really fun. By then, maybe I'll be playing on a well-built board as a participant rather than an administrator. I hope so.