Are you in a position to throw any development resources towards perl itself? Hugo's concern about maintainability is definitely warranted, as there's a limited amount of engineering resource available to perl, and past decisions have definitely made an impact on what's easy and what's hard with the current code base. The best way to make sure what you want to have happen actually happens is to help make it happen.
in reply to STOP Trading Memory for Speed
Update: This isn't meant to be snide or anything--many people are unaware of how much time and effort adding even seemingly simple features to a large and complex project like perl, nor how much time you have to spend afterwards maintaining that code...