Build log · visioncont.org

Vision Control

I started with a fairly simple observation.
Over the past 30 years, a huge number of billionaires have come from software. That makes sense. Software has high startup costs, but once it exists, the marginal cost of serving one more user is close to zero.
So I started thinking about what I could build. At first, I hired someone on Fiverr to make a prototype for $2,000. It wasn't terrible, but it wasn't especially good either. More importantly, it didn't really match what I had in my head.
That was when I started taking vibe coding more seriously. I know just enough about computers to understand where things are likely to break, but not so much that I need total control over every line of code. That turned out to be a pretty good position to be in.
I watched some YouTube videos about how to vibe code correctly, worked through the learning curve, and eventually reached the point where I could build what I was actually imagining. Once that happened, the idea became very clear.
I had a vision seared into my mind of what version control for non-technical people should look like. It shouldn't feel like GitHub. It shouldn't require people to understand branches, commits, repositories, or any of the other language that makes sense to developers but not to everyone else. It should simply show you what changed.
I scratched out a few ideas in a notebook, filed for a business license, and started building Vision Control. A lot of decisions had to be made along the way. Not just technical decisions, but product decisions. What should be visible? What should be hidden? How much detail is useful? What is the clearest way to compare two versions of something?
I also learned that vibe coding can quickly turn into a doom loop. You ask for one change. That change breaks something else. You fix that problem, and the fix creates another problem. The AI keeps producing code, but the product slowly drifts away from what you originally wanted. The only way around that is to know exactly what you want and be stubborn as an ox about getting there.
That idea became part of the philosophy behind Vision Control. I see Vision Control as a small pushback against AI slop. A lot of AI tools encourage people to generate something, glance at it, and decide it is good enough. But “good enough” is usually where the creative process starts falling apart. Vision Control makes you slow down and look at every difference. It makes you decide what improved, what got worse, and what still needs to change.
The goal is not to reject AI. I used AI to build the product. The goal is to stop the algorithm from taking control of your creative process. Use the tools. Keep the judgment.
Control your vision completely.