A lot of my side projects end up being: “I want to learn how to do X, but I have no idea what that’s going to take.” And I end up just coding a logical step a

A lot of my side projects end up being: “I want to learn how to do X, but I have no idea what that’s going to take.” And I end up just coding a logical step and if I don’t know what to do go research it, realize I did something wrong 3 steps ago and then either revert or abandon depending on my current motivation. I’ve tried before to create a “game plan” before but most of the time there’s so much that I don’t know that I don’t know that it’s either impossible to make that plan or once I do, that plan ends up being wrong anyways. As such it always feels like I’m doing side projects wrong. But since they’re more learning endeavors more than actually trying to launch a product maybe that’s fine?