What is SQL?

Structured Query Language — the universal grammar of relational data.

Visual explanation diagram · click steps to walk through it

Definition

SQL (Structured Query Language) is the ANSI/ISO standardised language for talking to relational databases. It dates back to IBM in the 1970s and is the lingua franca of every relational DB — MySQL, PostgreSQL, Oracle, SQL Server, SQLite all speak it.

How It Works

SQL is declarative — you describe WHAT you want (SELECT name FROM users WHERE age > 18), the database engine figures out HOW to get it. Plans, indexes, join order — all the database's job.

Example

sql
-- Your first SQL statement
SELECT 'Hello, SQL!' AS message;

Expected Output

(Run in MySQL Workbench, pgAdmin, or any SQL playground — actual values depend on the data in YOUR tables.)

Why It Matters

Every backend role on the planet expects SQL fluency. Learning it once pays dividends across every database vendor for your whole career.

Key Takeaways

Standardised by ANSI (1986) and ISO (1987); updated roughly every 5-7 years.
Declarative (what) vs imperative (how) — the engine plans execution.
Same SQL runs on Postgres, MySQL, Oracle, SQLite — modulo vendor extensions.
Three sub-languages: DDL (CREATE/ALTER/DROP), DML (INSERT/UPDATE/DELETE/SELECT), DCL (GRANT/REVOKE).
Read-mostly workloads = SELECT-heavy; write-heavy = INSERT/UPDATE-heavy.

Interview Questions

Practice Questions

Run SELECT VERSION(); on any SQL database — note the dialect.
Compare the same SELECT in MySQL and PostgreSQL — observe identical behaviour.
Read the Wikipedia entry on SQL standards — note ANSI vs ISO timeline.

Pro Tips

When learning, default to a standards-compliant subset — ports easily across DBs.
Watch out for vendor extensions (AUTO_INCREMENT, RETURNING, etc.) that aren't portable.
Books: 'SQL Performance Explained' (Markus Winand) is the classic.

// Try It Yourself · SQL Playground

in-browser SQLite · no network

Run any SELECT, INSERT, JOIN or GROUP BY against the seeded users · products · orders tables — your changes are sandboxed and reset whenever you click Reset DB.

// editor · Ctrl + Enter to run

// result

Click Run to see results here.

// see seed schema (3 tables · 8 users · 8 products · 12 orders)

users (id, name, email, country, age, created_at)
products (id, name, price, stock, category)
orders (id, user_id → users.id, product_id → products.id, qty, total, created_at)

// Challenges0 / 6 solved

Each challenge is auto-graded — write the SQL, click Submit answer, and we'll compare your result-set to the canonical one.