The abandoned side-project that succeeded by teaching its author what it was meant to do

A developer recounts building a Svelte + Netlify quiz app to drill Latvian noun cases — a language with seven cases, two genders, and roughly 84 possible noun endings to memorise. The stack was deliberately minimal: serverless functions for question delivery and answer checking, a static JSON noun list, local storage for high scores, and a regex-driven engine to strip stems and validate suffixes. After a week of evenings, the MVP shipped and immediately revealed its flaw — the quiz was trivial, and without three wrong answers it ran forever.

The twist: by researching conjugation patterns deeply enough to encode them, the author had internalised the rules he built the app to teach. The deliverable was obsolete on day one because the build process itself was the learning mechanism. He abandoned it the same day.

The broader argument pushes back on developer hustle-culture’s “always be shipping” framing and the tendency for tech recruiters to weigh candidates by side-project output. Abandoned projects aren’t failures by default — they’re prototypes, skill-building exercises, and scratch pads. Success is measured by what was learned, not just what was deployed, and treating side-projects as throwaway experiments removes the shipping pressure that makes them feel like obligations.