Right, but some people are doing data wrangling by themselves with python in a jupyter notebook. These aren’t production systems, and static type checking slows

Right, but some people are doing data wrangling by themselves with python in a jupyter notebook. These aren’t production systems, and static type checking slows you down without adding much value. That’s all I’m saying, the data analyst that has some edge case “bug” that never gets executed probably shouldn’t pay the upfront overhead cost of type checking.