Pseu, Pseu, Pseudio. Pseudonymization in Django.
summary
Demonstrates concrete pseudonymization implementations for Django: a first working prototype using properties plus a queryset monkeypatch to allow filtering on masked values, and a cleaner version implemented as a custom Django field that masks in get_prep_value and unmasks in from_db_value. Shows admin form adjustments, discusses tradeoffs (masking vs encryption, space and CPU costs), links to sample code, and explains where masking fails and when tokenization or encryption is needed.
AI-written summary · about
chapters (32)
- 0:00Intro
- 0:53Who is Frank
- 1:57Who is Phil Collins
- 3:16Spotify playlist
- 3:50What is Pseu
- 4:16Why would you do this
- 4:39What is personal data
- 6:36GDPR
- 7:14Anonymization
- 8:40Data Masking
- 9:09Approximation
- 9:40Encryption
- 10:43Tokenization
- 11:19Implementation Example
- 12:58Mask and Unmask
- 13:16Implementation
- 14:00Django example
- 14:14Quick question
- 14:32Shifting algorithm
- 14:43User attribute
- 15:42Updating the query set
- 15:57Filter and exclude
- 16:22Code
- 16:59Defer
- 17:33Query using Filter
- 17:49Updating Django Admin
- 18:06Defining a form
- 19:58New Improved Method
- 20:34Custom Fields
- 21:31Field Source Code
- 22:56Questions
- 23:11Packages
related talks
32mAnonymization and Dynamic Masking with PostgreSQL
50mPrivacy in Practice with Smart Pseudonymization: Lessons from the Belgian Public Sector
48mPractical Django Secuirty
29mPushing The Pony's Boundaries - Django Admin Customization
12mStriking the Balance: Safeguarding Customer Privacy While Empowering Employees
35mExternal authentication for Django projects
26mMaking Django Really, Really, Ridiculously Secure (CW)
34mDatabase Anonymization with PostgreSQL