in reply to
Re^5: Truly randomized keys() in perl 5.17 - a challenge for testing?
in thread Truly randomized keys() in perl 5.17 - a challenge for testing?
What is the justification for "simplifying" code that has worked perfectly well for a dozen or more releases?
Maintainability and testability. Why else would you simplify code?
You omitted to mention: "broke a bunch of other people code for no good reason"
It didn't break any code. It made existing bugs occur more often. That's actually a good thing!