Friday's notes:
Highlights:
- Dominik gave an awesome demo of the Racket Roguelike Library
- Brad (@Gambiteer ), author of SRFI 231, described at a high level a deforestation-like approach in generalized arrays
- Why bother deforesting
range
when it doesn't improve performance on its own? - Why do we still need
begin-encourage-inline
when we have already saiddefine-inline
? (we don't know. Any ideas?) - The huge potential of Qi's "no accidental side effects" policy
We also finally returned to the discussion about good ways to test the compiler, and adopted a clever technique suggested by @samth and @gus-massa to test the compiler's initial normalization pass. We're going to continue exploring this next time so you're welcome to join if you're interested in either sharing or learning more here.