The Eric Normand Podcast

How do we evaluate a data model?

10 snips
Oct 31, 2022
Ask episode
AI Snips
Chapters
Transcript
Episode notes
INSIGHT

Two Critical Steps in Data Modeling

  • A data model has two critical steps: creating a conceptual model and encoding it in code.
  • Problems arise if either the conceptual model is poor or the encoding doesn't fit that model well.
ANECDOTE

M.C. Escher Art Example

  • Eric uses the M.C. Escher art example to illustrate finding important concepts like tiling and flipping in images.
  • These concepts form a conceptual model that makes encoding and drawing art much easier than pixel-by-pixel manipulation.
INSIGHT

Fit Is Key in Data Models

  • The ideal data model encoding perfectly overlaps the conceptual model states with one-to-one correspondence.
  • It's often better to allow encoding invalid states and handle errors than to miss encoding valid conceptual states.
Get the Snipd Podcast app to discover more snips from this episode
Get the app