The article examines whether Julia can address scientific computing’s “two-language problem”: the common need to prototype in easy, slow Python and then rewrite performance-critical code in faster but less convenient languages such as C++ or Rust. It argues that this matters because Python’s usability has made it dominant, yet its speed limitations still impose extra work, complexity and barriers for researchers. By framing the issue through the history of programming languages, the piece presents Julia as an attempt to combine Python-like ease with much higher performance.

The article links this modern problem to earlier efforts to bridge human-friendly notation and machine-friendly programming, especially Kenneth Iverson’s APL, introduced 60 years ago in 1966. It notes that Julia was created in 2012 by four mathematically minded computer scientists, who described themselves as “greedy” for wanting a language that was open source, easy to learn and powerful enough for serious technical work. Their goal was to avoid the trade-off found in languages such as Matlab, Lisp, Python, Ruby, Perl, R and C, each of which they saw as strong in some areas but weak in others.