# Decision Record ## Decisions - Readiness is `L2 concept / pending-measurement packet`; this repo is not a shop packet. - Treat the jal tarang as a tuned vessel idiophone where water-volume repeatability is the central engineering problem. - Keep bowl material open: porcelain, ceramic, glass, and metal remain candidate branches until real bowls are measured. - Add a cut-list as candidate test/setup classes only, with all quantities, sizes, pitch targets, and water states still pending measurement. - Separate bowl inventory, fill workflow, pitch logging, mallet trials, support behavior, and session drift as independent evidence streams. - Do not create CAD, DXF, tuning tables, water-volume tables, pitch targets, acoustic predictions, or mallet specifications in this round. - Keep every artifact authority at `concept_only` or `pending_measurement`. ## Alternatives Considered - **Final scale first:** rejected because measured bowl response and performer context should govern any future scale choice. - **Permanent water-level marks now:** rejected because labels could imply tuning authority before repeatability data exists. - **One support tray design now:** deferred until bowl size, material, and resonance behavior are observed. - **Mallet specification now:** deferred until rim safety and attack tests exist. - **CAD tray layout now:** rejected because no bowl inventory or support geometry has measured authority. ## Open Questions - Which reference tradition, repertoire, or performer setup should anchor the next packet? - Should the first study use a graduated found-bowl set or a more uniform bowl family? - How should water volume be measured so tuning is repeatable after draining and refilling? - What mallet material gives clear attack without damaging bowl rims? - What support surface keeps bowls stable without damping them too much? - How should labels show bowl ID and fill state without implying final tuning? - How much evaporation or handling drift matters during a practice session? - What bowl condition notes are needed before deciding whether a candidate bowl is usable? - How should a measured fill state be reset after transport or rehearsal? - What evidence distinguishes true tuning drift from measurement-room or striking variability? ## Promotion Gates - L2 records subsystem intent, candidate part classes, and pending measurement gates only; it does not authorize fabrication or performance setup. - L3 requires measured bowl dimensions or reviewed drawings plus measured pitch-versus-water data and validated support behavior. - L4 requires physical setup logs, repeatability tests, player feedback, and documented tuning drift over time.