This whole week I got my database and API-powered set of Python scripts working together and testing lead me to tweak things, and then shit started to become top-heavy. By adding slightly more features, the app was having all sorts of problems and issues loading. The first version of the app ran off of a CSV file, which limited the automation and things I could do, so then, I stepped up to a SQLite database and an API, and learned about how SQLite handles threading... There's a reason why they call it SQLite!
Talking to the AI and planning things out, it was obvious. I needed to build a better foundation with what I've learned from the first two versions of the app.
So far, I migrated over to the plugin system and things are already much better, Python is real picky about loading shit in, and I had a major spaghetti load going on. But how would I know? "I'm just vibe codin', man!" (said in senile Joe Biden voice)
The 1st app I made, I got about 40% of the way through before the amateur coding methods made the app impossible to make. The 2nd app, I got about 90% of the way to a fully-running app before the programming limitations bit me in the ass. After talking more with ChatGPT and migrating with Claude, the much-improved programming practices like AppContext will create a much better base for scaling up without the app shitting its pants. As much as it sucks having to re-write your code over and over and over to get it just right, it is also fun to learn new things, especially when I think I'm close to having an app that will greatly increase my work output, and, hopefully, prevent me from starving!
Anyone here planning on any vibe coding today? I had a little puff of some weed... so nice, just you and the AI, the clicks on the keyboard, the taps of rain on the leaky tin roof, vibe coding in a shack... it's more fun than drinking!
