Agreed wholeheartedly. There are times when you should read and learn every detail before actually starting a new activity. (e.g. ground school before flying)

Agreed wholeheartedly. There are times when you should read and learn every detail before actually starting a new activity. (e.g. ground school before flying) This is usually when the cost of failure is unacceptably high. However, this is usually not the case, especially with things like programming, if that's what OP is talking about. Advice to "just do it" may seem simple, but it is surprisingly effective. When you learn by doing, you setup a feedback loop where you try something, see if it works or not, and if it does not, find out why/where it went wrong and retry it. For most people, this is mentally stimulating and provides the necessary psychological rewards to keep you going. If you get stuck/frustrated, this is when you should take a break and perhaps go back to reading/researching after a while. Don't worry about whether you're doing it the absolutely correct way, or whether what you've done is worthwhile to others. (What is "absolutely correct" is often subjective, anyways) Focus on the knowledge you gain, and take notes. For example, I've also been recently learning Golang. I wrote a simple Golang game server (using a SockJS library) to serve as a backend for a simple, multiplayer online game. Looking back, the initial code I wrote was horrible and needs to be refactored; but that in and of itself will be another learning exercise. Did I need to use Golang for this, or is this the best use of Golang? Probably not - I could have written it using Java or Node.js much faster, but wouldn't learned as much.