I'm curious, why add another operator?

I've not been party to the discussion to implement this, so take my take on it in that light.

Basically, some -- I would say most -- of the time, you do not want this behaviour.

Most of the time, if you have a variable that should contain a reference, that your program logic requires you to call a method on (or dereference to use as a hash or array ref) then if it does not contain a reference, it is an error that your program cannot safely continue from, and all you can do is stop with an error message. Ie die.

But there are sufficient cases where a potential reference not having a value is not fatal; that having to code an explicit test before each dereference clutters what could otherwise be clear and concise code.

The use of different operators allows the programmer to cater for both situations cleanly, with little fuss.