We have it backwards.
Replace NULL with explicit semantics. Use -999 for "offline," -9999 for "out of range," or better—split the column into value and value_metadata_flag . 3. The Referential Integrity Illusion Modern data lakes love "schema on read." This is the enemy of ab initio . You are essentially saying, “Let’s store the garbage, and we’ll figure out what kind of garbage it is later.” ab initio data quality
Most data teams focus on reactive data quality (DQ). They let data in, then scramble to fix it. But what if we borrowed a concept from theoretical chemistry and quantum physics? What if we focused on ? We have it backwards
Use tools like pydantic (Python), Great Expectations (with expect_column_values_to_not_be_null set to fatal ), or dbt 's constraints (enforced, not just documented). If the contract fails, the pipe breaks. Loudly. They let data in, then scramble to fix it