Okay, I did a little bit of research on my own to add to what you have given me... But what is the difference between these two snippets of code?:

---

boo_radley's way, using @{$db{'key1'}{'key2'}[0][0]} = "one_value";, produces this hash:

%db = (
         'key1' => {
                     'key2' => [
                                 [
                                   ['one_value'],
                                   ['two_value']
                                 ],
                                 [
                                   ['three_value'],
                                   ['four_value']
                                 ]
                               ]
                   }
       );
[download]

---

mstone's way, using $db{'key1'}{'key2'}[0][0] = "one_value";, produces this hash:

%db = (
         'key1' => {
                     'key2' => [
                                 [
                                   'one_value',
                                   'two_value'
                                 ],
                                 [
                                   'three_value',
                                   'four_value'
                                 ]
                               ]
                   }
       );
[download]

---

The only difference between the two is that referencing it as an array (@{$db{...}}) adds square brackets around the values. But what difference does this create in reading/storing values?

(Hash printout via Data::Dumper)

UPDATE: So, I should use mstone's code?