Suppose you have a slice declared as:

names := []string{"Manyanda", "Abhishek", "Pawel", "Eoin", "Jameel", "Vincent", "Jason", "Massimo"}

and a printSlice function that prints the first count element of the passed in slice implemented like this:

func printSlice(slice []*string, count int) {
    for i, val := range slice {
        if i < count {
            fmt.Println(*val, " rules!!")
        } else {
            return
        }
    }
}

Now suppose that to invoke it you use the following code:

// convert names to a slice of string pointers
var ary []*string
for _, val := range names {
  ary = append(ary, &val)
}

// Let's print all the names but the last one (Massimo)
printSlice(ary, 7)

The output of the previous code will be:

Massimo  rules!!
Massimo  rules!!
Massimo  rules!!
Massimo  rules!!
Massimo  rules!!
Massimo  rules!!
Massimo  rules!!

Explanation

For performance reasons, go reuses the loop variable instead of creating a new one on every loop. By adding the address of the loop variable to the slice on each loop, we always added the same address and the variable value will be the last value it has been assigned in the loop (Massimo).

The fix is easy:

// convert names to a slice of string pointers
var ary []*string
for _, val := range names {
  tmp := val
  ary = append(ary, &tmp)
}

// Let's print all the names but the last one (Massimo)
printSlice(ary, 7)

Now the output will be:

Manyanda  rules!!
Abhishek  rules!!
Pawel  rules!!
Eoin  rules!!
Jameel  rules!!
Vincent  rules!!
Jason  rules!!