Skip to content

0120-synthetic-data-privacy.md

Latest commit

 

History

History
36 lines (26 loc) · 1.48 KB

0120-synthetic-data-privacy.md

File metadata and controls

36 lines (26 loc) · 1.48 KB
Raw

ADR 0120: Synthetic Data Generation and Privacy Compliance

  • Status: Proposed

Context

In the development and testing of the TicketsPlease system, we need realistic data sets to verify UI behavior, performance, and business logic. Traditionally, developers sometimes use anonymized production data, which carries significant legal and security risks under the General Data Protection Regulation (GDPR / DSGVO).

Decision

We will use the Bogus library to generate 100% synthetic data for all non-production environments (Development, Testing, Staging).

  1. Strict Locale: Data generation will use the de (German) locale to ensure that addressing, names, and postal codes are realistic for our target market.
  2. Infrastructure Hook: The seeder will be integrated into the Infrastructure layer and triggered via the Web entry point.
  3. No PII: No real "Personally Identifiable Information" (PII) will ever be used or imported from external sources for seeding purposes.

Consequences

  • Compliance: Full compliance with Datenschutz (GDPR) requirements as no real user data is involved.
  • Independence: Developers can reset the environment at any time with a fresh, populated data set.
  • Consistency: All developers work with similar data distributions.
  • Maintenance: The DbInitialiser must be updated whenever the domain schema changes.

References