As one of the original Perl Hackers I agree that this irrational Perl Hate Minute stuff is silly. To some extent language preferences are religious, but that doesn't mean we need a holy war! My response to the "I emigrated penniless from my home country rather than use Perl, now you want to profane Zope, I will immolate myself" wailing is, frankly, g_e_t__a__l_i_f_e. Perl was born as a slack hack but it's grown up and you can produce great, clean, readable apps in it with a modicum of effort. You can also knock off some kinds of quick and dirty real world applets in a few (non-obfuscated) lines, where a "purer" OO language needs several non-obvious paragraphs and/or advanced tricks. On the other hand, you can write bad Perl and give people headaches. You can also do this in Python. Will the availability of Perl methods mean that fewer people learn Python? Probably, but it will be a smaller percentage of a bigger user "pie." My primary worry (if you're listening, Digicool) is that integrating the Python and Perl runtime environments on various platforms will lead to unexpected conflicts and performance problems. For example, will threading be enforced properly? If memory leaks, whose memory is it? If Perl releases a new version, will Zope admins need to hold back until it's cleared for use with Zope?
(Note: I pulled a Guido, making this announcement before going on the road for two days. I've spent about three hours compiling a blanket response, but it seems that I have another two more before I'm done!) Tom Neff wrote:
My primary worry (if you're listening, Digicool) is that integrating the Python and Perl runtime environments on various platforms will lead to unexpected conflicts and performance problems. For example, will threading be enforced properly? If memory leaks, whose memory is it? If Perl releases a new version, will Zope admins need to hold back until it's cleared for use with Zope?
Thought I'd try to address this one specifically. The first two are listed explicitly in the contract with ActiveState. That doesn't mean there can't still be weird interactions. Jim Fulton has debugged a lot of ZODB corruption problems caused by memory errors in, of all things, a MySQL DA. Translation: vigilance. --Paul
participants (2)
-
Paul Everitt -
Tom Neff