Right — that’s why I’m calling it source‑available. I’m mainly testing user behavior around distribution, not trying to define what’s ‘open’ or not.

I’m not trying to prove which license is better — too many variables, like you said. I’m just testing how different models change user behavior: who clicks, who downloads, who ignores. It’s more about distribution patterns than software quality.

For this project I’m mainly testing distribution models. My only restriction is redistribution — people can read and modify the code for personal use. I’m also cautious about someone copying or commercializing it, so this is mostly a learning exercise for me.

Source‑available still lets people read and modify the code for personal use — they just can’t redistribute it. For me that’s a reasonable model for small tools, even if there’s always a risk someone will copy it. This project is mainly a distribution experiment.