August 9, 2026
Code Closed? Comment War Open
"Coding is solved" misses the point
AI can spit out code fast, but the comments say the real mess was never the typing
TLDR: The article argues AI is making typing software easier, but the real challenge in big companies is understanding rules, history, and business needs. Commenters split hard: some say that proves “coding was never the whole job,” while others insist AI will remake business itself and bulldoze those old constraints.
The article’s big claim is a buzzkill for the “AI solved coding” victory lap: yes, chatbots can crank out working software faster than ever, but real companies don’t run on clean little demo projects and vibes. They run on old decisions, team turf wars, security rules, hidden assumptions, and that classic office mystery: “why wasn’t this done already?” In other words, the hard part often isn’t typing the code. It’s figuring out what should be built, who owns it, and what breaks when you touch it.
And wow, the community did not quietly nod along. One camp basically yelled, “You’re thinking too small!” with johnwheeler arguing that today’s messy corporate rules may simply get bulldozed by a new wave of scrappy AI-built companies. Another group went full semantics police, with 9rx pushing back on the article’s framing and asking whether “coding” ever meant anything beyond turning a clear request into software. Then came the veteran energy: commenters warned that more code is not automatically better, because in older systems, extra code can be less treasure and more ticking time bomb. Even the humble “add a button” got roasted, thanks to a nod to the classic XY problem, aka solving the wrong thing confidently.
The funniest mood? A mix of hype, eye-rolling, and battle-scarred realism. One commenter even compared it to saying “chess is solved,” which is nerd shorthand for: people are declaring victory way too early.
Key Points
- •The article defines "coding is solved" narrowly as the growing ability of language models to turn well-specified problems into runnable code.
- •It argues that enterprise software work is constrained by organizational context such as service ownership, existing APIs, security rules, architecture, and business priorities.
- •The article says software engineering includes multiple layers: business objective, product design, solution design, and implementation.
- •It claims AI is reducing the cost of implementation and moving the primary bottleneck to higher-level decision-making.
- •The article places language models in a historical pattern alongside compilers, higher-level languages, and frameworks, each of which automated a concrete layer of development.