Description
The usual compromise in numerical work is to prototype in Python because it is quick to write and slow to run, then rewrite the hot loops in C or Fortran. Julia tries to remove that step: script-like syntax with speed close to a compiled language.
Its core idea is multiple dispatch: one function name can behave differently for different combinations of argument types, which suits math-heavy code well. It ships a package manager and can call C, Fortran and Python libraries, so old code does not have to be rewritten.
It is MIT licensed. The official download page has installers, a portable ZIP that needs no installation, and an LTS release.
Command-line REPL: Work interactively from the terminal, or run scripts.
Multiple dispatch: Define behavior by the combination of argument types.
Near-compiled speed: A dynamic language designed for technical computing.
Calls other languages: Use C, Fortran and Python libraries directly.
Package manager: Built in, for installing and managing packages.
Multiple builds: Installer, portable ZIP and an LTS release.
Its core idea is multiple dispatch: one function name can behave differently for different combinations of argument types, which suits math-heavy code well. It ships a package manager and can call C, Fortran and Python libraries, so old code does not have to be rewritten.
It is MIT licensed. The official download page has installers, a portable ZIP that needs no installation, and an LTS release.
Features
Command-line REPL: Work interactively from the terminal, or run scripts.
Multiple dispatch: Define behavior by the combination of argument types.
Near-compiled speed: A dynamic language designed for technical computing.
Calls other languages: Use C, Fortran and Python libraries directly.
Package manager: Built in, for installing and managing packages.
Multiple builds: Installer, portable ZIP and an LTS release.
