The Seven Sins of Personal-Data Processing Systems under GDPR

Supreeth Shastri, Melissa Wasserman, Vijay ChidambaramHotCloud 2019 · OpenReview

TL;DR

This paper argues that modern computing systems, built with a focus on scalability and performance, inherently conflict with the principles of the General Data Protection Regulation (GDPR). The authors identify seven "sins" – common design practices that violate GDPR – and discuss the challenges of retrofitting existing systems for compliance.

What Problem Does This Paper Address?

In recent years, a surge in privacy and security breaches has highlighted the need for stronger data protection. The European Union's GDPR, enacted in 2018, aims to address this by establishing comprehensive regulations for personal data processing. However, the paper points out a fundamental disconnect: the way modern computing systems are designed and operated often clashes with GDPR's requirements. This creates a significant challenge for companies trying to achieve compliance, especially given the deep-rooted nature of these design principles. The paper seeks to bridge this gap by analyzing these conflicts from a systems design perspective.

Methods

The authors employ a qualitative analysis approach. They review GDPR regulations, specifically focusing on articles and recitals that pertain to data processing principles and individual rights. This is then juxtaposed with common design principles, architectural components, and operational practices found in modern computing systems, particularly those prevalent in the Internet era. The core of their method is to identify and categorize these conflicts into seven distinct "sins," illustrating each with real-world examples and discussing the implications of GDPR's regulations on these practices.

Datasets

Not discussed in the paper.

Key Insights and Ideas

  • The "Seven Sins" Framework: The paper introduces a novel framework of seven "sins" that encapsulate common system design practices conflicting with GDPR. This provides a structured way to understand the challenges of GDPR compliance from a systems perspective.
  • Inherent Conflict: The paper highlights that the very goals that made modern systems successful (scalability, performance, cost-efficiency) are often at odds with GDPR's emphasis on privacy, transparency, and user control.
  • "Storing Data Forever" is a Major Hurdle: The practice of retaining data indefinitely for potential future use or analytics directly contradicts GDPR's "right to be forgotten" and "storage limitation" principles. The paper explains how data replication and distributed systems make timely deletion a significant technical challenge.
  • Purpose Limitation vs. Data Fungibility: GDPR's strict "purpose limitation" principle clashes with the common system design approach of treating data as a fungible resource that can be reused across various services.
  • GDPR as a Catalyst for Change: While challenging, GDPR is forcing a re-evaluation of how systems are built, pushing for more privacy-by-design approaches and greater transparency.

Key Results

The paper doesn't present quantitative results in the traditional sense. Instead, its key findings are qualitative insights into the systemic conflicts between GDPR and modern computing practices. The "results" are the identification and detailed explanation of the seven GDPR sins:

  1. Storing Data Forever: Conflicts with the "right to be forgotten" and "storage limitation."
  2. Reusing Data Indiscriminately: Conflicts with "purpose limitation" and the need for specific user consent.
  3. Walled Gardens and Black Markets: GDPR's "right to data portability" and information requirements for data procurement aim to dismantle these opaque data ecosystems.
  4. Risk-Agnostic Data Processing: GDPR mandates "Data Protection Impact Assessments" and "prior consultation" for high-risk processing, directly opposing a "move fast and break things" mentality.
  5. Hiding Data Breaches: GDPR enforces mandatory, timely notification of breaches, contrasting with historical practices of concealment.
  6. Making Unexplainable Decisions: GDPR's "right to access" and the concept of a "right to explanation" challenge the black-box nature of many automated decision-making systems.
  7. Security as a Secondary Goal: GDPR's "data protection by design and by default" principle elevates security and privacy to primary design considerations, not afterthoughts.

Limitations and Future Work

The authors acknowledge that labeling design choices as "sins" is inherently controversial, as systems were not designed with future regulations in mind. They also note that many of these practices are driven by technical and economic realities, and achieving compliance may lead to performance overheads and slower product rollouts.

Future work suggested includes exploring compliance strategies at different levels (infrastructure components vs. individual regulations), developing better testing methodologies for compliance (proactive vs. reactive, automated vs. manual), and determining how compliance costs should be factored into cloud pricing. The paper also points out that while they focused on seven systematic violations, other issues like online behavioral tracking also warrant attention.

Summary updated

BibTeX
@inproceedings{shastri2019seven,
  author = {Supreeth Shastri and Melissa Wasserman and Vijay Chidambaram},
  title = {{The Seven Sins of Personal-Data Processing Systems under GDPR}},
  booktitle = {HotCloud},
  year = {2019},
  url = {https://openreview.net/forum?id=dNxRHxmGuj},
}
Figure