Simple Made Inevitable: The Economics of Language Choice in the LLM Era (felixbarbalet.com)

A recent analysis emphasizes the importance of language choice in developing with large language models (LLMs), particularly highlighting the advantages of Clojure in minimizing accidental complexity. The piece recounts the author's experience managing microservices at Qantas and outlines how frameworks like Polylith help tackle the inherent "entropy" of growing codebases. The discussion references Fred Brooks' distinction between essential and accidental complexity, asserting that LLMs are particularly adept at managing the latter, yet they can also generate additional complexity when not properly guided. Significantly, the author argues that as LLMs become more integrated into software development, the barriers associated with learning new languages diminish. With this shift, criteria such as developer familiarity are being overshadowed by the intrinsic simplicity of a language’s abstractions. Clojure stands out for its token efficiency, allowing LLMs to handle larger context windows efficiently, thus reducing the overhead of complex boilerplate code through its immutable data structures. This suggests a paradigm shift in language selection where simplicity and efficiency become paramount, especially for large-scale projects, ultimately prompting a reevaluation of how developers choose programming languages in an era dominated by AI-driven development.