Episode: 56

Data Driven Domains: We're all wrong about SQL

This week we go full data modelling nerd and ask a simple question with messy consequences:

Are you designing your domain model, or are you letting your database (and ORM defaults) design it for you?

We cover:

  • Model-first vs database-first: why your domain should lead, and the code smells when it doesn’t
  • Relational reflex: the habit of assuming SQL is always “right” (and when it absolutely is)
  • Aggregates and boundaries in DDD and why document stores often map more naturally than tables
  • Many-to-many headaches in ORMs (junction tables, cascades, deletes) and how they leak into your domain
  • Normalisation vs denormalisation: performance, reporting, and the trade-offs you’re actually making
  • A quick tour through non-relational models hierarchical, object, even just indexed files, and what they get right
  • Practical ways to keep persistence concerns out of your domain and avoid “entity = table” thinking

It’s opinionated, occasionally ranty, and very much a “save future you from your schema” kind of episode.

🍻 Tonight’s Drinks
Liam: Glandore Hunter Valley Shiraz (Sorry I forget what year it was. But it was good) 🍷

Matt: Nothing this evening

Any Likes 👍, Shares 📣, Subscriptions 🔔, and Love ❤️ help us keep this purely-for-fun thing rolling.

Cheers! 🍻