This article follows how I developed an implementation of the Graham scan algorithm in Haskell, missteps and all. I think it’s valuable to see the process that others use, especially in a “weird” language like Haskell. At times, it can seem like most of what you’ve learned about developing software doesn’t apply to Haskell, but I think that’s just because the language allows so much of the scaffolding to be cut away once the software is finished. Reading Haskell code written by the masters can feel like looking at the Sistine chapel and wondering where they got such long paint brushes. This will be a step-by-step implementation as I developed it. I’m still learning too, so don’t take this as an example of excellent Haskell style.

Read the rest…