This was intended to be friendly, helpful and non-competitive.

Perhaps you could've replied in a similar vein along these lines:

---

Thanks for taking the extra time to compare benchmark results. However, further testing on the intended (platform/OS/Perl version) indicates that my version will run faster and therefore I will be going with that solution.

---

Instead of:

• Using an unsubstantiated claim about what I understand. Are you psychic?
• Including pointless rhetoric. Neither you nor I claimed your (or my) code was faster. Perhaps take a moment to consider why you added this.
• Taking my statements out of context.
  • Yet again you have focussed on passing array vs. arrayref while ignoring what is happening in the closure.
  • Claiming I was talking about speed when I was, in fact, talking about speed differences (the former is absolute, the latter is relative).

Finally, I very much doubt that anyone able to understand the closure generation code will have the slightest problem dereferencing a scalarref. As such, I believe your initial target for improving user-friendliness should be the closure.