A row is one single record in a table — the values for every column, describing one specific thing.
Why This Matters
- Almost every operation in SQL — insert, update, delete, and most of what SELECT returns — happens at the level of rows.
- Being precise about what a row actually is (and isn't) avoids a surprising number of beginner mistakes later in this course.
Core Concept
- A Row (formally, a Tuple) is one entry in a table — an ordered set of values, exactly one per column.
- Every row in a table has the same columns as every other row, in the same order, even though the values differ.
- "Row" is the everyday term; "tuple" is the formal, set-theory term — the same relationship as Table/Relation.
Deep Dive
- A row represents one instance of whatever the table models — one customer, one order, one product — not a category or a type. The table "customers" models the concept of a customer; each row is one actual, specific customer.
- Rows in a well-designed table are meant to be independently meaningful and independently changeable — deleting or updating one row should never require touching any other row just to stay consistent (you'll see exactly what breaks this rule, and how, in the Normalization section).
Common Mistakes
Designing a table where deleting one row accidentally destroys information that belonged to a different, unrelated row.
Why: This usually means two different kinds of facts got squeezed into one table, so a row is doing more than one job.
Fix: If deleting a row could lose information you still needed, that information probably belongs in its own table — a preview of what the Normalization section covers formally.
Interview Questions
What is a row, formally?
A tuple — an ordered set of values, one per column, representing a single record in a table.
Chapter Summary
A row (or tuple) is one record in a table — one specific instance of whatever that table represents. Every row shares the same columns as every other row in its table, but holds its own independent values.