Data Modelling in Practice
A schema is a set of promises.
A course on modelling data for systems that must be queried, migrated and trusted years later. It covers normalisation, temporal data, identity, soft deletion and the reporting disasters that begin as innocent shortcuts.
4.7h
Study time
13
Lessons
4
Modules
4.8
Rating
203 reviews
Enrolment
Enrol to open Data Modelling in Practice in the learning workspace, track progress and take the assessments.
Demo enrolment — no payment or real enrolment occurs. Sign in to see the full workspace.
Subjects covered
5 topics
Knowledge map
What you will be able to do
- Model a domain so that invalid states are hard to represent
- Handle time, history and correction without losing the truth
- Design identifiers that survive integration
- Anticipate the reporting questions a schema will be asked
Curriculum
4 modules · 13 lessons
- 01Entities, Relationships and LiesA close look at normalisation and the failures it is meant to prevent.24m
- 02Making Invalid States UnrepresentableMaking identity explicit enough that somebody else could repeat it.26m
- 03Normalisation and When to StopWhere temporal data usually goes wrong, and how to notice it early.22m
- 01Natural Keys and Their BetrayalsHow migration works in practice, once the deadline is real.20m
- 02Surrogate Keys Done ProperlyWhere reporting usually goes wrong, and how to notice it early.18m
- 03Identity Across SystemsMaking normalisation explicit enough that somebody else could repeat it.22m
- 01Valid Time and Transaction TimeThe working method for identity, reduced to what actually matters.26m
- 02Append-only ThinkingHow temporal data works in practice, once the deadline is real.22m
- 03Soft Deletion and Its CostsA close look at migration and the failures it is meant to prevent.20m
- 04Auditing Without DrowningMaking reporting explicit enough that somebody else could repeat it.18m
- 01Migrations as a Product RiskThe judgement calls behind normalisation, and what each of them costs.24m
- 02Backfilling SafelyThe working method for identity, reduced to what actually matters.22m
- 03Designing for the Report You Have Not Been Asked ForThe judgement calls behind temporal data, and what each of them costs.20m
Assessment
Practical submissions marked against stated criteria, plus low-stakes module quizzes.
Quizzes
No quizzes in this course — assessment is by written submission and exercise.
Assignments
No assessed assignments — each lesson closes with an unassessed exercise.
What students say
- 5 out of 5Rowan Petersen — Product designer
The reading is short and the practice is not, which is the right way round. I finished the first module in an evening and then spent a week on the exercise.
- 5 out of 5Samuel Okonjo — Researcher
I have taken a lot of courses on this subject and this is the first one that explains why the rules exist rather than just listing them.
- 5 out of 5Silje Haugen — Analyst
The quiz explanations are worth taking seriously — I learned more from the two I got wrong than from the four I got right.
Fictional reviews — demonstration content only
Related courses
- 014.9out of 5 from 298 reviews
Architecture Under Pressure
Designing systems for the day they fail.
Advanced6.9 h18 lessonsTomas Ekelund — Principal Engineer in Residence
- 024.7out of 5 from 611 reviews
Frontend Engineering Foundations
The browser, honestly explained.
Beginner4.9 h13 lessonsTomas Ekelund — Principal Engineer in Residence
- 034.8out of 5 from 412 reviews
Modern Product Design
From ambiguous brief to defensible interface.
Intermediate7 h20 lessonsRafael Otero — Senior Lecturer in Product Design