Because scientific computing requires the highest performance, most science related libraries are written in C/Fortran and define a high-level API, most commonly in Python. Julia solves this “two-languages” problem, and also offers many more benefits. In my talk I want to focus on something that I consider a big, but often unstressed, asset of Julia: the fact that it brings unprecedented code clarity and intuition, both of which are crucial for scientific progress. I want to argue about how Julia removes “black-boxes” and “blind-trust” by allowing you to easily inspect and understand source code without being a developer. The packages of the JuliaDynamics GitHub organization (currently: DynamicalSystemsBase.jl, ChaosTools.jl and DynamicalBilliards.jl) have been written to take full advantage of this asset of Julia. In my talk I will briefly overview them and show examples of how one can have a 1-1 correspondence between computer code and scientific thought and algorithms.
