>>>>> "KAG" == Kenneth A Graves <kag-K3RN/aI3HbloiYX5Tdu9fQ@xxxxxxxxxxxxxxxx>
>>>>> writes:
KAG> On Fri, 2006-10-20 at 11:42 -0400, Mitchell N Charity wrote:
>> >We also
>> >learned that the Perl6-ish Moose.pm, while making it much more convenient
>> >to write object code, adds a significant amount of overhead to each method
>> >call, which is prohibitive for a non-interactive script like Kenneth's
that
>> >does a huge number of method calls.
>>
>> Moose.pm is said to now be significantly faster, in part as a result of
>> Kenneth's experience. Though the improved version has apparently not yet
>> reached CPAN (the one there dates from just before the PM meeting).
KAG> My presentation was done with Moose 0.12. The latest version in
KAG> FreeBSD's ports is 0.13, to which I've just upgraded. Slower. (Factor
KAG> of 25 more user time instead of 15 with 0.12.) Moose 0.14 is on CPAN,
KAG> but doesn't have anything about speed improvements in Changes. (There
KAG> are some mentions about optimizations they hope to make in the POD.)
KAG> I'm guessing the above refers to 0.15 or later.
at the yapc::chicago hackathon i hung out with stevan little, one of the
major moosebrains out there. i helped him find a decent optimization of
a slow point. given my basic skimming of code and that module, i suspect
there are plenty of ways to speed up moose. they seem to be focusing on
the functionality first. seems like an interesting idea but i will stick
to my blessed hashes for a little while longer (like when p6 is in
stable release! :).
uri
--
Uri Guttman ------ uri-X/06uaNR9nFmbZtjAW+qKA@xxxxxxxxxxxxxxxx --------
http://www.stemsystems.com
--Perl Consulting, Stem Development, Systems Architecture, Design and Coding-
Search or Offer Perl Jobs ---------------------------- http://jobs.perl.org
|
