I think you've hit the nail on the head: execution speed is apparently
a sore spot in the Ruby community, one that results in undue
defensiveness and shooting the messenger! From my reading, Perl and
Python are the primary competitors to Ruby among programming languages
at this time. I prefer the elegance of Ruby, and want it to be chosen
over the others when potential users are making such a decision.
Whereas Python cannot change its indentation requirement (and remain
Python) and Perl (version 5) cannot change its messiness, I believe Ruby
is capable of reaching a comparable level of execution speed. Solving a
problem is usually better accomplished by acknowledging it openly rather
than causing people to feel like they have committed heresy or stupidity
by raising it.
Regards,
Jamal
···
--Original Message--
From: gregarican [mailto:greg.kujawa@gmail.com]
Sent: Friday, July 14, 2006 4:00 PM
To: ruby-talk ML
Subject: Re: How to speed up ruby and make it as fast as possible
There are certain sore spots in language discussions which tend to get
feathers ruffled. Seeing this thread has garnered so many replies in
such short time uncovers a sore spot in Ruby language discussion. Ruby
tends to be slower than other similar interpreted scripting languages.
Although some folks won't admit it as a broad generalization this
assertion tends to be stereotypically correct.
Every language has its sore spot. For Python try bringing up
significant whitepace or passing the understood 'self' around each and
every method. For Java try bringing up the truth of 'write once run
everyone' or mixing primitives in with objects for a supposedly OO
language. For Perl try bringing up cryptic lack of readibility and
"bolted on" version updates. For PHP try bringing up secure coding
practices. And so on...
I don't think that's it. The common speed complaint on this list stems from someone looking at some benchmark or another. They see Ruby places behind language X, so they complain about it.
But performance is really far more complex than that. For example, consider PHP4. On paper, with benchmarks, it is faster than Ruby, and simple benchmarks put it much faster than Ruby for the things PHP is good at -- dynamic web content.
But...go download a PHP MVC framework like CakePHP (which is a pretty nice piece of PHP). Now compare it to a Ruby framework. Suddenly the benchmark numbers start changing, dramatically. My tests with IOWA show a speed difference of about 4x Ruby's favor when comparing CakePHP under PHP4 to IOWA under Ruby 1.8.4. So which language is truly faster?
I think a person is hard pressed to find an application which will perform acceptably under Perl or Python, but which will not under Ruby.
So while additional speed is nice, I have yet to be convinced that it is the big deal that some people keep trying to make it.
Kirk Haines
···
On Sat, 15 Jul 2006, Jamal Mazrui wrote:
Python) and Perl (version 5) cannot change its messiness, I believe Ruby
is capable of reaching a comparable level of execution speed. Solving a
problem is usually better accomplished by acknowledging it openly rather
than causing people to feel like they have committed heresy or stupidity
by raising it.
I think you've hit the nail on the head: execution speed is apparently
a sore spot in the Ruby community, one that results in undue
defensiveness and shooting the messenger! From my reading, Perl and
Python are the primary competitors to Ruby among programming languages
at this time. I prefer the elegance of Ruby, and want it to be chosen
over the others when potential users are making such a decision.
Whereas Python cannot change its indentation requirement (and remain
Python) and Perl (version 5) cannot change its messiness, I believe Ruby
is capable of reaching a comparable level of execution speed. Solving a
problem is usually better accomplished by acknowledging it openly rather
than causing people to feel like they have committed heresy or stupidity
by raising it.
The real point is: Speed is only a problem if you think it is.
I've used Ruby for more than six years, and have never had a
problem with speed.
What annoys me is people getting in my face and saying that
I'm "ignoring the problem." I'm not. For my purposes so far,
there isn't a problem.
Everyone, nyself included, would like Ruby to be faster.
But I don't condone alarmists running around squawking about
their own problems as if they were everybody's problems.
Some people find Ruby "unusable" because of its speed. Those
people should use something else until Ruby is fast enough.
(And/or they should contribute to making Ruby faster.)
However: Many, many people don't find Ruby "unusably slow."
As I said, I have never had an application where Ruby's
speed was a real problem. (Once, four years ago, I had an
app that took several minutes to run. But it needed to run
very rarely. If I were running it several times a day, then
it might be an issue.)
So in short, anyone is welcome to consider Ruby "slow" in
a given context. But don't say that it's "slow" in every
context, or that there is a "problem" that I'm "ignoring."
The real point is: Speed is only a problem if you think it is.
Matz thinks it is. He's said so elsewhere in this thread.
I've used Ruby for more than six years, and have never had a
problem with speed.
It's good that you have not had this problem.
What annoys me is people getting in my face and saying that
I'm "ignoring the problem." I'm not.
If someone asks "this Ruby script is too slow for my purposes, how can
it be faster" and is only told "write it in C", I think they can
reasonably say that *their* problem is being ignored.
Some people find Ruby "unusable" because of its speed. Those
people should use something else until Ruby is fast enough.
(And/or they should contribute to making Ruby faster.)
Which is why I asked if YARV is the way to go. If I knew that I could
see if I could help. But (/me sighs) nobody answered that. Anyway
I've just downloaded and benchmarked YARV and it's very impressive.
Many times faster on most benchmarks, merely comparable on a few.
If anyone has tips on helping YARV, I'm all ears.
Regards,
Jeremy Henty
···
On 2006-07-14, Hal Fulton <hal9000@hypermetrics.com> wrote: