Executive summary: Perl is awesome, Java sucks, and there are also some other programming languages.   This certainly fits with my experiences.  They’ve been a bit more scientific and pedantic.  Enjoy.

Shared with: Public, Paul Fernandez

+1’d by: Brandy Butler, tony d, Paul Fernandez

Comments

Paul Fernandez - 2013-07-17T04:39:28Z

As a Perl lover, this is great. As a skeptic, I wonder if this is all on the level. He seems to have pro-Perl/anti-Java bias…Chicks, did you write this? :)

Christopher Hicks - 2013-07-17T13:58:40Z

True greatness is measured by how much freedom you give to others, not by how much you can coerce others to do what you want.

-- Larry Wall

Christopher Hicks - 2013-07-17T14:15:32Z

+Paul Fernandez : I understand your concern.  Java’s reputation has strongly benefited from a well-oiled hype machine so it is uncomfortable when that pig is so easily gored.  I don’t think any of that is a surprise to anyone that has used Java outside the classroom.The other languages get a fair shake.  LUA is not aimed at solving these sorts of problems so the results make sense.  Python did better than I expected in performance, but considering the way he defined his benchmark he’s not using many of pythons sexy slow features so I can believe that part.  Of course PHP is shit.  I hadn’t heard of magic quotes before.  That’s exactly the sort of short-sited ill-conceived thinking that I’ve bitched about from the PHP/FI community. Then there’s Perl.  People have been unfairly beating it down for speed since the beginning.  How could a bunch of interpreted code run fast?  Well, that’s parsed and thrown out in the first millisecond.  Afterward you’ve got the core Perl event loop executing tokens.  Those tokens are like an assembly language that provides cool features like a regex engine that is shockingly efficient.  And they’ve been optimizing Perl5 for over a decade.  Features show up too, but there has been a lot of debugging, optimization, trying it out and letting that feed back into the process.  The result is shocking: your Perl code runs faster than equivalent C code.  Of course you could write your C code with all of the crazy optimizations that are already built into Perl.  But I’d rather not reinvent the wheel so I code mostly in Perl.And I did not write it.  You can prove that by word usage patterns. You have plenty of what I’ve written publicly and privately to make a corpus.  :)  I with the guy had edited it more thoroughly.  I’m tempted to redo it myself and include more obscure languages.  I’ll post it here if I do!