Scaling over decades
Scaling over decades
CSV is viable long-term if proactively managed. To keep repositories performant and diffs focused, BusDK supports splitting data into multiple files by time period or category. A typical strategy is to segment journal entries by period using root-level files with a date prefix, such as journal-2025.csv and journal-2026.csv, instead of allowing a single file to grow indefinitely. The repository root tracks these files through journals.csv, which records which periods exist and where each file lives. This reduces the size and complexity of day-to-day diffs, keeps Git operations snappy, and aligns with the practical expectation that even large datasets can remain manageable when partitioned. Older data can be archived by tagging year-end commits, and where desired, by removing old-period files from active branches while retaining them in history for retrieval.