fig. 2.9.1: every convention the code assumed, counted over the 70-project corpus and ranked by the audit fig. 2.9.1 · every convention the code assumed, counted over the 70-project corpus and ranked by the audit.⛶ THE AUDIT CAUGHT THE GATE’S CONVENTION IN 3 OF 69 REPOS ONE ROW, ONE CONVENTION; BAR = SHARE OF CORPUS; ACCENT = DEFECT FIXED 0% 50% 100% A2 · COMMIT SUBJECTS MATCH ^Session N 3 OF 69 CRITICAL · CANON NEVER RESOLVES, THE JUDGE NEVER RUNS A1 · CONTENT FILES END IN .md 25 OF 70 SEVERE · INDEXES NOTHING, REPORTS NOTHING WRONG A6 · WRITE FRONTMATTER + CANON TAGS 0 UNMODIFIED REPOS MIGRATION REQUIRED BEFORE INDEXING A3b · PROPER NOUNS MATCH [A-Z][a-z]+ A REGEX, NOT A COUNT MODERATE · ACCENTED NAMES MISSED OR TRUNCATED A5 · AN elaborations/ DIRECTORY 37 OF 70 MODERATE · MIGRATION NO-OPS A5 · A map FILE 36 OF 70 MODERATE · MIGRATION NO-OPS A4 · THE PROJECT IS A GIT REPO 69 OF 70 HOLDS A3a · FILES ARE UTF-8 400 OF 400 HOLDS PROJECTS P AND I SATISFIED FOUR OF THESE AT ONCE. EACH WAS TESTED AGAINST THE 70-PROJECT CORPUS.
source: cairn DESIGN.md, assumption audit, lines 762-808 and 810-836 (repo 12dd72d) · assumptionAudit.data.ts
View in context: Counting the assumptions a test shares with its code → · All figures
Andrew Fribush , software and computer engineer in Austin. I build consequential systems where software meets institutions: finance, healthcare, commerce, government data, and the mechanisms that govern AI agents. Taking new engagements: hello@andrewfribush.com .