Fortran turned 70 on October 15th, 2026, dated from the publication of its first manual, and it remains the load-bearing wall of scientific computing. Not legacy code kept on life support — active, evolving infrastructure that climate models, nuclear simulations, and weather forecasts depend on daily. The Institute of Computing for Climate Science at Cambridge is throwing it a proper celebration. The event lands on December 1st, 2026 at Manchester's Science and Industry Museum, which is exactly where you'd want it: a building that houses industrial history, including 'Baby,' the 1940s experimental computer that will be demonstrated during the evening reception. There's a pleasing symmetry in celebrating a language born in the punch-card era alongside hardware from its grandparent generation. The workshop runs 14:30–18:15 with a keynote from Damian Rouson of Lawrence Berkeley National Laboratory, lightning talks on current Fortran developments, and breakout sessions on the next 70 years. Registration is free, remote attendance via Zoom is available, and the deadline is November 9th, 2026. Lightning talks and posters are open for submission. What makes Fortran's persistence remarkable is that it isn't nostalgia. Python dominates data science. Rust is eating systems programming. Julia was designed specifically to replace Fortran and C in scientific computing. None of them have. Fortran's array-handling semantics and compiler optimization for numerical computation remain unmatched at the scale climate and physics codes demand. The language has modernized through successive standards — Fortran 90, 95, 2003, 2008, 2018, 2023 — adding modules, object-orientation, coarrays for parallelism, and modern memory management while keeping backward compatibility with code written decades ago. The event is organized by Cambridge's ICCS, which sits at the intersection of computing research and climate science — the exact domain where Fortran's dominance is most absolute. Their funding model means attendance is free, lowering the barrier for early-career researchers and community members who might otherwise skip a niche workshop. The cross-posting to the CAKE website signals the broader computational science community's interest. This is less a review of a cultural artifact and more a marker of an ongoing phenomenon: a programming language that predates integrated circuits, survived every paradigm shift in computing, and remains indispensable. The celebration is warranted not because Fortran is old, but because it's still working.
