Why are you so hung up on this?
It seems to be more or less a matter of taste.
There are plenty of CPAN modules that take a list of raw name/value pairs,
and plenty that take a hash reference instead.
As pointed out by ikegami, the way Moose does it is probably more efficient.
OTOH, Perl Best Practices chapter nine ("Use a hash of named arguments for any subroutine that has more than three parameters")
advises "as tempting as it may be, don't pass them as a list of raw name/value pairs".
Conway recommends passing named arguments as a hash reference instead,
because, as ikegami also noticed,
doing so gives you a good chance of catching
common blunders, such as the one made by the OP,
at compile-time ("Odd number of elements in hash"), rather than run-time.
Update: As indicated at Named Subroutine Parameters: Compile-time errors vs. Run-time warnings, this "reason" was an error in the book: anonymous hash population is done at run-time, not compile-time. Conway still stands by the advice though because "Error messages that point users to the right place are definitely worth the (tiny) overhead of passing named args in a hash".
BTW, PBP chapter nine used to be the free sample chapter, but has now become a broken link. :-(
