Making Programs Faster

Pseudo-Hashes

It was all very complicated
- Lots and lots of code had to be added to Perl
- All sorts of complications
- exists had to be extended to work on arrays
After it was all done, however, the new improved semantics were 15% faster than the old:

        Old                     New

                             package Critter;
                             use fields qw(... hearts ...);

                             my Critter $self;

        $self->{hearts};     $self->{hearts};