CSE Roadmap

Specialist tier · about 2 hours · 8 parts

Part 7 of 8 · 2 min read

Privacy beyond law

Think about the person, not only the rule.

Privacy beyond law

DDIA ends with a warning. It repeats a warning that data is the pollution of the information age, and that history will judge how societies contain and dispose of it. It notes the 1995 European Data Protection Directive: personal data must be collected for specified, explicit and legitimate purposes, and be adequate, relevant and not excessive. DDIA doubts the law is fully effective, since it runs against the "collect everything and explore" approach. It also says over-regulation could block benefits, such as better medical diagnosis, so the balance is hard.

For an engineer, the useful habits are: ask what each field is for before collecting it, keep a map of where personal data lives, restrict who can see it (module 7), and make deletion a built-in feature, not an emergency.

BM adds a human point from the same chapter: think about processes for people leaving, and imagine what a disgruntled former employee could do with the access they still hold.

Interview angle

Answer in your head first, then open it.

Where does cloud cost come from?

Compute, storage, network and managed services, plus the observability data you keep. Network and forgotten storage often surprise people.

How do you cut cost safely?

Right-size, scale with load, use cheaper tiers and keep traffic local, as long as the error budget allows.

How do you design for the right to be forgotten?

Keep a map of where personal data lives, use a purge flow that reaches every store, expire backups, and test it end to end.

What is data frugality?

Keep only what you need for a stated purpose.

Check yourself: 1. What does a data map help with? 2. Why does DDIA doubt the law alone is enough? 3. What should happen to access when someone leaves?

Words you will meet

Word Meaning
Cost driver A quantity that makes the bill grow
Right-sizing Matching machine size and count to real load
Tagging Labelling resources by team and service for cost split
Unit cost Cost of one useful unit of work
Personal data Information about, or able to identify, a person
Data frugality Keeping only the data that is needed
Retention rule How long a kind of data is kept
Excision Truly removing data from an append-only history
Purge Deleting one user's data across all systems