Schema

Describes how data is structured

Conceptual schema: in terms of the Data Model
Physical schema: in terms of how it's stored on disk

To create a schema, we need to do Requirement Analysis

• What data does the application collect?
• What data does this application need?
• What does the application do with the data?

Is a DBMS the right fit?

• Big data, ML, timescale DBs...

In relation to Agile Development

• Refactoring / restructuring wrong data choices is pain (high chance of losing data)

Key design principles:

• Don't store computed variables, but raw, unchanging ones (eg birthdate instead of age)
• Names (and most string-based fields) are hard.