May 10, 2026
7 lines, infinite opinions
7 lines of code, 3 minutes: Implement a programming language (2010)
A tiny code trick blew minds as commenters argued if this is genius or just showing off
TLDR: A classic article says you can build a tiny but surprisingly powerful programming language in minutes, turning a scary topic into something almost playful. Commenters split between calling it essential learning for everyone and mocking it as a neat trick done in a language that already makes the job easy.
A 2010 post claiming you can build a whole programming language in 7 lines of code and about 3 minutes has sparked exactly the kind of reaction you’d expect online: awe, eye-rolls, nitpicking, and a little nerd poetry. The article itself is basically a love letter to the idea that making your own language teaches you how computers really think. It uses an ultra-minimal system of functions and shows how something that looks tiny can still be powerful enough to run forever, crash your brain, or both.
But the real show is in the comments. One camp was utterly charmed, calling the exercise a “lovely experience” because it forces you to decide which ideas matter and which are just clutter. Another went even bigger, arguing this kind of thing should be part of everyone’s education, right alongside building a simple computer from blinking lights, because modern life runs on software whether you’re buying a croissant or casting a vote.
Then came the skeptics. One commenter delivered the thread’s sharpest side-eye: this is basically “implement lambda calculus in languages that are pretty much lambda calculus,” which is internet shorthand for: Congrats on doing the easy version in a language already built for it. Another commenter went full gotcha mode on the JavaScript example, posting error messages like a courtroom exhibit and asking, essentially, “Am I missing something, or is this wrong?” Meanwhile, one fan casually revealed the article has been their X banner for a decade, which is either wholesome devotion or peak programmer fandom depending on your mood.
Key Points
- •The article presents a 7-line interpreter for a Turing-equivalent functional programming language based on lambda calculus.
- •It includes three implementations: a 7-line Scheme version, a Racket reimplementation, and a 100-line interpreter with additional language features.
- •The article describes lambda calculus as having only three expression types: variable references, anonymous functions, and function calls.
- •It explains that Alonzo Church developed lambda calculus in 1929 and that it was later shown to be computationally equivalent to Alan Turing's Turing machine.
- •The article uses examples such as the identity function and the non-terminating Omega expression before introducing the interpreter as an environment-based denotational interpreter in R5RS Scheme.