This is great post and the observation is quite applicable in pretty much every subject that you want to learn and master, not just Python. In fact, "learning t

This is great post and the observation is quite applicable in pretty much every subject that you want to learn and master, not just Python. In fact, "learning to learn" is perhaps the most important skill we develop as humans and a small improvement in this area translates to order of magnitude improvement in actual learning. In PhD programs they say that getting PhD is not actually about mastering specific subject but rather developing an ability to comb through vast amount of knowledge, assimilate it and synthesize new knowledge. The first two steps is where much of the effort lies which is "learning to learn". The first mistake that most people make in first step is trying to understand everything cover-to-cover of each paper before moving on to next. An experienced person would first try to get big picture (just look at the abstract and results :)) and move on to next one. So you do breadth-first-search followed by some selective depth-first-search and continue back and forth until you build "surface" that represents the usable approximation of state of the vast amount of knowledge. This is many order of magnitude efficient than try to build most of the "surface" from the get go.