architecture-journallisted

# Architecture Journal > "Architecture decisions are the most important decisions we make — and the ones we most often fail to record. The cost of a forgotten decision is not the decision itself, but the weeks spent relitigating it." > -- Michael Nygard, "Documenting Architecture Decisions" ## Core Philosophy Experience without reflection is just time passing. This skill converts architecture decisions into transferable expertise through structured recording and timed retrospectives. The goal is not documentation — it is building judgment. **The Journal Principle:** A decision recorded at the moment of choosing, reviewed at 30, 90, and 180 days, teaches more than a decade of unexamined experience. ## Domain Principles | # | Principle | Rationale | Violation Signal | |---|-----------|-----------|------------------| | 1 | **Record at Decision Time** | Capture reasoning when it is fresh, not months later when memory has rewritten history | ADRs written after implementation; reasoning reconstructed from code | | 2 | **Context Over Conclusion** | Why matters more than what; the decision is visible in code, but the reasoning is invisible | ADRs that state what was chosen but not what forces drove the choice | | 3 | **Lightweight Over Comprehensive** | A recorded decision beats an unrecorded perfect analysis; 15 minutes of writing beats 0 minutes | Blank ADR directories; teams that "never have time" to write decisions down | | 4 | **Retrospective at Fixed Intervals** | 30/90/1