Rubinius X is no more

Received this mailer:

Hello ,

Thank you for signing up to the Rubinius X mailing list.

I want to share with you some important news about the project: I’ve decided to retire the idea of “Rubinius X” as a separate project, but not the goal of providing a powerful way to extend your investment in Ruby while taking advantage of the changing environment for building applications and Software as a Service.

If you wish to unsubscribe from this list, I’ll be disappointed because I have some exciting stuff to share with you soon, but I’ll understand. Otherwise, I’ll be merging this list with the main Rubinius project mailing list.

[If you have a few minutes, please take our survey: “Who’s Using Ruby (or Not), for What, and Why?”

Before you go, I’d like to explain a bit more about Rubinius X…

If you’re like me, you probably use Ruby more than a few times a week and really appreciate it for what it gets right, but are really frustrated by what needs to improve. Nearly 2,000 people have signed up to this list, so we’re not alone.

When I first described my ideas around Rubinius X, I had spent many years writing thousands of tests for Ruby behavior and implementing a lot of it. The things I described, like better data structures and more powerful modularity support, are still desperately needed in Ruby.

However, over the past year and a half that I spent improving the Rubinius infrastructure to support ideas in Rubinius X, I’ve come to accept two things: 1. a powerful platform to enable as many people as possible to experiment with language ideas is critically important; and 2. I no longer think there is value in trying to improve Ruby itself.

The goal of Rubinius has always been to provide a better technical foundation for Ruby. With the generational garbage collector, LLVM-based JIT compiler, and no global interpreter lock, we’ve built a solid foundation. We intend to maintain this fundamental support for your investment in Ruby while focusing most of our effort on significant alternatives to Ruby.

For example, many Rubyists are using Go to get better concurrency, speed, and easy deployment. But as a developer, I don’t believe you should be forced to accept a trade-off between those features and a managed environment for quickly prototyping applications. One of the capabilities we’re building into the Rubinius platform is the option to run a program on the virtual machine or export it to a machine code executable.

That’s just one example. If you’re interested, videos from a couple talks I presented this year are available below.

“The Next 10 Million Programmers” at So Coded 2015:
“The End Of General Purpose Languages: Rubinius 3.0 And The Next 10 Million Programs” at Philly Emerging Tech 2015

Finally, if you haven’t had a chance to read my series of posts on Rubinius 3.0, here are some handy links. Almost all the infrastructure work is complete and I’m excited to start delivering Rubinius 3.0 features soon.

“Rubinius 3.0 - Part 1: The Rubinius Team”
“Rubinius 3.0 - Part 2: The Process”
“Rubinius 3.0 - Part 3: The Instructions”
“Rubinius 3.0 - Part 4: The System & Tools”
“Rubinius 3.0 - Part 5: The Language”

I’d love to hear from you. Drop me a note with your comments or questions.



Another update:


One of the biggest challenges we have with Rubinius is knowing where to focus resources to provide the greatest value to all of you using Rubinius. We’re experimenting with ways to close the feedback loop. One of the simplest ways is with surveys. Today, I want to ask you a couple questions about a code analysis service.

Recently, Code Climate announced their platform, “the first open and extensible platform for static analysis”. A feature of their platform is the specification for writing “engines” that they will run when they analyze your code. See

One of the things we’ve focused most on in Rubinius is building the infrastructure for running your code in a way that can support sophisticated tools. Rubinius generates and uses extensive runtime data. We could use this data together with a custom Code Climate engine to extend their analysis in powerful ways that depend on actual program behavior. Alternatively, we could build a separate Rubinius service, one aspect of which could be code analysis.

We’re curious which one you would find the most valuable. Please take a few minutes to fill out our survey:

Also, If you haven’t yet had a chance to fill out our other recent survey, Who’s Using Ruby (or Not), For What, and Why? we’d really appreciate hearing from you.

We recently posted on the blog about our new versioning scheme, release process, and using Gitter to improve project communication:

“Distributed Coding, Distributed Releases”
“MAJOR.MINOR: Maximize Delivering Features, Minimize Trouble”
“Rubinius :heart: Gitter IM”

If you had signed up for the Rubinius X mailing list, I recently posted to that list about the status of the project and merging that list with this list. Email if you have any questions.

We always love to hear from you. Drop us a note


1 Like