tl;dr:“B-trees are typically slower than hashmaps, but I was surprised to find fairly wide variation in people's expectations of how much slower.” Jamie compares the speed of b-trees and hashmaps.
tl;dr:The following allowed Jamie to program faster: (1) Caring and wanting to be faster. (2) Making decisions based on goals. (3) Focus in blocks of 2-3 hours i.e. no email, slack, twitter... (4) Batching tasks. (5) Make small changes for big tasks. (6) Shorten feedback loops. (7) Write stuff down. (8) Reduce frequent mistakes. (9) Make low-level skills automatic. (9) Reflect. Each is discussed with examples.
tl;dr:Things Jamie used to believe that he's changed his mind about: (1) There is actually a simple solution to a problem, and "everyone is doing it wrong." (2) Programming should be easy. (3) You should always search for the new thing. (4) Learning X will make you a better programmer. (5) Intelligence trumps expertise.
tl;dr:Jamie believes there is "a huge amount of value to be unlocked by replacing SQL, and more generally in rethinking where and how we draw the lines between databases, query languages and programming languages.