Data Design — Restoration Decision
Data Design — Restoration Decision
Historical decision; superseded by the worked example at
docs/helix/02-design/data-design.md.
data-design is restored as the canonical design-activity artifact for schema,
storage, access-pattern, and migration design.
Decision
This artifact is restored rather than retired.
Current HELIX still names data-design in the design README, state machine,
and design gate. That intent is also not replaced cleanly by architecture,
contracts, or security documents.
Why It Exists
- Architecture explains the system shape and component boundaries.
- Contracts define external interface and schema obligations.
- Security architecture defines trust boundaries and controls.
- Data design defines the internal data model, storage strategy, lifecycle, access patterns, and migration consequences that implementation and tests must follow.
Canonical Inputs
docs/helix/02-design/architecture.md- feature-level solution or technical designs when a slice owns specific data changes
- relevant contracts when external schemas constrain internal storage
docs/helix/01-frame/compliance-requirements.mdanddocs/helix/02-design/security-architecture.mdwhen retention, classification, or protection rules shape the model
Minimum Prompt Bar
- Identify the entities, stores, and key fields that matter to the design.
- Make relationships, cardinality, and lifecycle consequences explicit.
- Document the main read/write access patterns and their performance needs.
- Record integrity, retention, privacy, and classification constraints that materially affect the model.
- Define migration and rollback expectations for schema or storage changes.
- Stay at the design level instead of drifting into ORM or query implementation details.
Minimum Template Bar
- data summary
- entities and stores
- relationships
- access patterns and constraints
- validation and security consequences
- migration strategy
Canonical Replacement Status
data-design is not replaced by architecture, contracts, or
security-architecture. Those artifacts provide context and constraints, but
data-design remains the canonical place for data-model structure and
evolution strategy.