Re: Perl Advent Calendars 2009

[stvn]

Too bad the Perl Advent calendar has gotten off to a bit of bumpy start, the author seems to have misread the Mouse docs and made some incorrect assumptions.

It (Mouse) runs 4 times faster than Moose according to its authors, at the expense of implementing only 96% of its larger cousin's features.

Mouse runs a subset of the Moose test suite in 1/4 of the time, mostly due to startup overhead and the recent XS work done by Goro Fuji (which will eventually be ported to Moose, he is currently using Mouse as a testing ground for these ideas).

As for the "96% of the features" part, nothing could be further from the truth. The Mouse docs say "Mouse code coverage is also over 96%", not that it implements 96% of Moose's feature set. Mouse is basically Moose without the MOP which is where Moose gets most of its power and flexibility and without which most of the MooseX:: modules could not be written (which is why you don't see comparable MouseX:: out there).

Too bad the author didn't run his post by the Mouse/Moose community to check his facts.