REST Limitations & GraphQL Benefits

In 2012, Facebook faced a crisis. Their mobile app was slow, buggy, and draining batteries. The culprit? Their REST API.

The app needed multiple requests to render a single news feed item - one for the post, one for the user, one for comments, one for likes. Each request returned more data than needed. On slow 3G connections, the experience was terrible.

So Facebook invented GraphQL.

The Over-Fetching Problem

Imagine building a user profile card. You only need name and avatar. With REST:

// REST: GET /api/users/123
{
  "id": 123,
  "name": "Sarah Chen",
  "email": "sarah@example.com",
  "avatar": "https://...",
  "phone": "+1-555-0123",
  "address": {
    "street": "123 Main St",
    "city": "San Francisco",
    "zip": "94102"
  },
  "preferences": { ... },
  "createdAt": "2020-01-15",
  "lastLogin": "2024-01-20"
  // ... 20 more fields you don't need
}

You wanted 2 fields. You got 30. That's wasted bandwidth and slower parsing.

With GraphQL, you ask for exactly what you need:

# GraphQL Query
query {
  user(id: 123) {
    name
    avatar
  }
}

# Response - exactly what you asked for
{
  "data": {
    "user": {
      "name": "Sarah Chen",
      "avatar": "https://..."
    }
  }
}

The Under-Fetching Problem

Now imagine a blog post page. You need the post, author info, and comments. With REST:

// Request 1: Get the post
GET /api/posts/456
// Response includes author_id: 123

// Request 2: Get the author
GET /api/users/123

// Request 3: Get comments
GET /api/posts/456/comments
// Response includes commenter IDs

// Requests 4-13: Get each commenter's info
GET /api/users/201
GET /api/users/202
// ... and so on

With GraphQL, one request gets everything:

query {
  post(id: 456) {
    title
    content
    author {
      name
      avatar
    }
    comments {
      text
      author {
        name
        avatar
      }
    }
  }
}

How GraphQL Works

GraphQL is a query language for your API. Instead of multiple endpoints, you have one endpoint that accepts queries.

Real-World Adoption

GraphQL vs REST: Quick Comparison

Key Takeaways

Next up: Schema & Types - Learn how to define your data model in GraphQL.