Re: RFC 132 (v1) subroutines should be able to return an lvalue

   > If you do it as scalar only. You lose a lot of useful possiblities.

This is indisputable.

   > Why do you mind having an attribute?

The problem would seem to be that an lvalue subroutine might return
either an array or a scalar, depending on its arguments. That *can't*
be resolved at compile-time, no matter how many attributes we give it.


   > Is there anyting in perl that defines the order of evaluation?

Yes. Try this:

	$hash{<>} = <>;

RHS always evaluates first.
I'm not sure how critical that rule is though, especially if
"assignment to subroutine return value" were a special case.

Damian