{"body":"Today, Hyperliquid Taker Aggression values for Bitcoin, Ethereum, and Solana disappeared from one dashboard block. A storage check found all 96 fifteen-minute objects for the closed day.\n\nThe problem was in the data reader. It looked for a prepared daily file and, when that file was absent, treated the source itself as absent. This created an uncomfortable state: the primary data was available, but the final metric was empty.\n\nWe changed this route. If the daily file is missing, the reader now assembles the day from the exact fifteen-minute objects. The check returned complete coverage of 96 out of 96 windows for Bitcoin, Ethereum, and Solana.\n\nWe also fixed a separate interface rule. A real value remains visible under low coverage. If even part of the source-backed data exists, the dashboard shows the calculated value and adds a separate coverage-quality warning. An empty field means that a numeric observation truly does not exist.\n\nThe change is small. It is fundamental for an analytical product. A missing prepared aggregate, incomplete coverage, and missing source data are three different states. If they are collapsed into one, the user sees emptiness where a useful, though sometimes incomplete, signal actually exists.\n\nToday's failure came from a blind spot between storage and display. After the change, that boundary is explicit and verifiable.","content_language":"en","day_counter":{"label":"Day 359","number":359,"start_date":"2025-09-11"},"edited_at":"2026-09-04T12:58:20+00:00","id":"2026-09-04-when-data-exists-but-the-system-cannot-see-it","project_date":"2026-09-04","published_at":"2026-09-04T12:57:41+00:00","retrospective":false,"schema_version":"crash_brief_public_blog_post_v0_1","slug":"when-data-exists-but-the-system-cannot-see-it","source_language":"ru","source_types":["accepted_technical_history"],"status":"published","summary":"A dashboard value disappeared even though all source observations existed, revealing a blind spot between storage and display.","title":"The Blind Spot Between Data and the Dashboard","topic":"How Crash Brief separates missing aggregates, partial coverage, and missing source data","type":"problem_to_fix","updated_at":"2026-09-04T12:58:20+00:00","version":2}