1.1 · Tables, Rows, Attributes

Goal: master the core vocabulary of the relational model.

The relational model in one diagram

┌─── Table (relation) ─── Student ───────────────────────┐
│ id       name          class    dob          score      │
│ 1001     Alice Chan    F.4A     2007-05-12   86         │  ← Row (tuple / record)
│ 1002     Bob Wong      F.4A     2008-03-22   72         │
│ 1003     Carol Yip     F.4B     2007-11-08   91         │
└─────────────────────────────────────────────────────────┘
         ▲                                       ▲
         │                                       │
       Attribute (column / field)             Attribute

Core terms

Term	Meaning
Entity	A thing we model (Student, Book, Order)
Relation / Table	A collection of rows sharing the same attributes
Tuple / Row / Record	One specific instance of the entity
Attribute / Column / Field	A property of the entity
Domain	The set of valid values for an attribute (e.g. age 0..150)
Schema	The blueprint of the database (what tables, what columns, what types)
Instance	The actual data in a database at a moment in time

Why "relational"?

Relations between entities are represented by shared values — typically a primary key in one table appearing as a foreign key in another.

Student              Class
┌──────┬──────┐      ┌──────┬───────────┐
│ id   │ class│      │ id   │ teacher   │
├──────┼──────┤      ├──────┼───────────┤
│ 1001 │ F.4A │ ───▶ │ F.4A │ Mr. Lee   │
│ 1002 │ F.4A │ ───▶ │ F.4A │ Mr. Lee   │
│ 1003 │ F.4B │ ───▶ │ F.4B │ Ms. Wong  │
└──────┴──────┘      └──────┴───────────┘

Properties of a relation

A proper relation:

• Has a unique name.
• Has distinct attribute names.
• Stores atomic values (no lists inside a cell — that's 1NF).
• Has no duplicate rows (in theory).
• Row order does not matter.
• Column order does not matter in theory (though SQL preserves it).

Worked example · Library

Entity	Attributes
Book	isbn (PK), title, author, copies
Member	member_id (PK), name, class, joined_on
Loan	loan_id (PK), member_id (FK), isbn (FK), loan_date, due_date, return_date

Each row of Loan represents one borrowing event linking a member to a book.

Common student mistakes

• Storing comma-separated lists in a cell ("Alice, Bob, Carol") — violates 1NF.
• Mixing data types in one column.
• Duplicating data across many rows that should live in a related table.

Exam-style question

Q (4 marks): A school wants to track members of its many clubs. Each student may join multiple clubs and each club may have multiple students. Identify the entities, key attributes, and the relationship between them.

Sample answer:

• Entities: Student, Club, Membership (associative).
• Primary keys: Student.student_id, Club.club_id, Membership.(student_id, club_id) composite.
• Relationship: Many-to-many between Student and Club, resolved via the Membership associative table holding the two foreign keys plus optional fields like joined_on.

Key takeaways

• Relations are tables of tuples; attributes are columns.
• Atomic values, no duplicate rows.
• Relations between entities use shared keys.

➡️ Next: 1.2 Keys