D1 Migration

## Setup: Track this exercise with Beads

```bash
bd create "Praxis: D1 migration for [feature]" --type=task --labels=learn,infrastructure
bd create "Design minimal schema (question every column)" --type=task --labels=learn
bd create "Write up migration" --type=task --labels=learn
bd create "Write down migration (reversible)" --type=task --labels=learn
bd create "Add indexes with justification" --type=task --labels=learn
```

---

Help me design a D1 database schema for [FEATURE] following MY minimal data philosophy.

I'm building MY approach to database design. For [FEATURE]:

1. **Question every column**:
   - What data is truly required?
   - What can be derived instead of stored?
   - What's the minimal viable schema?

2. **Design the schema**:
   - Create table definitions
   - Apply proper types (TEXT, INTEGER, REAL, BLOB)
   - Add NOT NULL only where essential
   - Use sensible defaults

3. **Write migrations**:
   - UP migration (creates the schema)
   - DOWN migration (reverts cleanly)

4. **Index strategy**:
   - Which queries will be common?
   - Create indexes with clear justification
   - Avoid over-indexing

5. **Document MY data principles**:
   - When do I store vs. derive?
   - What's my naming convention?
   - How do I handle relationships?

Feature to model: [YOUR_FEATURE]
Expected queries: [YOUR_COMMON_QUERIES]