Go: What is the difference between nil and empty slice?

Nil slice — declared but never initialized; pointer, length, and capacity are all zero. Empty slice — has a backing array allocated but contains no elements.

Both behave the same for len(), cap(), range, and append. The key distinction appears in serialization and equality: a nil slice marshals to JSON null, while an empty slice marshals to [].

1var nilSlice []int             // nil
2emptySlice := []int{}          // non-nil, empty
3madeSlice := make([]int, 0)   // non-nil, empty
4
5fmt.Println(nilSlice == nil)   // true
6fmt.Println(emptySlice == nil) // false
7
8nilJSON, _ := json.Marshal(nilSlice)     // "null"
9emptyJSON, _ := json.Marshal(emptySlice)  // "[]"

Safely appending to a nil slice works — Go allocates the backing array on the first append. This means you can build slices from nil without special cases.

1var s []int
2s = append(s, 1, 2, 3)  // works perfectly
3fmt.Println(s)          // [1 2 3]

When to use:

• Nil slice — default state for a slice that will be populated later; natural zero value.
• Empty slice — when returning an empty collection in JSON or when you need slices.Equal(nilSlice, emptySlice) to be false.

Common mistakes:

• Comparing slices with == against a non-nil empty slice — only nil comparison works; use slices.Equal for content.
• Returning a nil slice from an HTTP handler — JSON response shows null instead of [], confusing clients.
• Not realizing append on a nil slice is safe — unnecessary if s == nil checks add noise.