It is an honor to have you comment on my post. You and your compatriots have done an excellent job with JRuby. In my opinion, your work is a model of efficient, effective open source development. I am grateful to the JRuby team.

You are correct, there is room on the JVM for JRuby, Groovy, and other languages.

I apologize if it seemed like I was trying to scare people away from J/Ruby. I was just offering my opinion, based on my experience, to Java devs who hadn’t tried either language yet. It’s it only with reluctance that I made my suggestion, because I really do appreciate the beauty of Ruby.

I haven’t tried JRuby since 1.1. I’ll take your advice and try it again, especially to see how it integrates with existing Java libraries.

-devdanke

Ruby 1.9 (or to be more specific the YARV implementation of it) will allow compilation to bytecode as far as I know. But this will be a JIT (just-in-time) compiler, so you will still distribute the source-code, and the virtual machine will optimize it to bytecode when running a piece of code for the first time. Pre-compilation of Ruby-code is not an easy task (or nearly impossible), as Ruby is very dynamic and allows changing all it’s code during runtime. Please correct me anyone if I’m wrong here.

Good article; a nice overview of some of the benefits. Too bad it got hijacked.

I found JRuby to be very useful to embed in an application to provide scripting support. Because of the ease of building simple DSLs in Ruby, I can provide a scripting interface that non-programmer users can basically understand, but that calls down into the apps original Java code.

It adds about 8MB to app download (after stripping rspec, rake and gems out), and startup is a little slow, but with some judicious threading, that can be hidden in other startup times.

I’m really happy with how JRuby has come along.

– Morgan

JRuby, despite being the fastest implementation and the most sensible deployment platform for a web app (assuming your app is multithreaded there is no contest) it has this stigma of being second class in some way. Or a trojan horse. Headius has already written about this in the elephant. This reputation is not deserved. JRuby will not kill Ruby, or destroy the community.

And neither is the canonization of Perl 6 as the Prime Vapor. People have been putting a lot of work into a system designed from the ground up to be both expansive and inclusive. When Parrot 1.0 is out and high level language development truly stabilizes it will be the best platform for writing Ruby 2.0. Matz could literally cherrypick from the Perl 6 featureset, do his magicawesome-syntax-fu on those enhancements, and voila Ruby 2.0 takes less than a year (after the date 1.9 is implemented). That said, Parrot 1.0 isn’t here yet so I can (sort of) understand why neither Matz nor the Rubinius team chose it.

My point is: stop using Perl 6 as the language that never made it. It is coming, it will be here, and once it is, a lot of things will change. And everyone who loves dynamic languages is going to cream themselves when they realize what we’re getting in Parrot. Sorry to jack the thread, I just see this meme so often and today devdanke’s comment was the final straw. Apologies for the bizarre rant but they say venting is healthy, right?

Viva la Ruby!