We received a lot of positive feedback on our LFortran announcement. Most generally like the idea and tool, and expressed interest to hear a bit more on why we think Fortran is a superior language in its domain and why it makes sense to use Fortran for new projects.

Why Fortran?

Fortran was designed from the ground up to naturally and simply translate mathematics to code that compiles and runs at maximum speed. And being specifically targeted for such fundamentally computational tasks, it contains a broad range of key functionality within the language itself, standard across all platforms, with no need for external libraries that may or may not be well optimized or maintained, at present or down the road.

We recently open sourced LFortran, an interactive Fortran compiler built on top of LLVM that we have been developing for the last 1.5 years. It is a work in progress and at the link you can find what works already, what is planned and a roadmap.

Here is our motivation.

Why are almost no new scientific or engineering software projects started in Fortran?

Usually (in our neck of the woods) C++ is chosen instead. The following are the most cited reasons for such a choice: