Tinkering wins most of the time. Honestly, unless it's a completely new concept or way of thinking (or for a lack of a better term, "academic" concept), tinkeri

Tinkering wins most of the time. Honestly, unless it's a completely new concept or way of thinking (or for a lack of a better term, "academic" concept), tinkering and building stuff will be faster and more fun. I only read books after I've tinkered enough to think "I can build something with X, but let's find out the best ways to do it". There are some problems with books for X language or Y framework: * A lot of books (especially free ones) have little to no exercises, giving you no chance to solidify your knowledge and instead copy and paste (or type along) * People that love reading books instead of tinkering, in my experience, a) take too long to learn b) don't retain any material beyond two weeks or so because they don't do exercises (if there are any) c) think they learned the material because they have it in short term memory The above also applies to a lot of "video courses" as well. The only good book I've read in the past years is http://haskellbook.com/ http://haskellbook.com/ because it explains things well, is well structured, and most importantly it is full of exercises.