I'm having trouble adopting this new LLM regime. What happens when a developer you rely on produces bad code one time out of ten? How about one time out of a hundred, but it's a severely critical bug? Who's going to find these issues after I'm gone?

Too many projects I wish I had time for. * A (new) impl of Python, Go, or Rust-like language for JVM. * Ruby library for emitting Project Babylon CUDA code. * Ruby LSP plug-in for JRuby Java integration. * New JRuby SSL library based on JDK crypto support. I need a team.

The Vector API for Java will have its 11th "incubator" release in JDK 16. Starting to wonder if we need Daenerys Targaryen to carry it into a funeral pyre before that egg will hatch. 🐲🐣

JDK 26 has been branched from the main line and is now in Rampdown Phase One: mail.openjdk.org/pipermail/jd... openjdk.org/projects/jdk/26/ #JDK26 #OpenJDK #Java

Now what do I do? 😩

This is what happens when you torture a prickly pear with barely enough sun and water to survive.

Why not use Java's instanceof patterns to null-check values in the middle of an `if` statement?

Somebody tell me why I shouldn't use instanceof pattern matching in Java to perform a null check plus variable declaration and assignment all at once: if (obj.getValue() instanceof Value value) { // use value } else { // it's null }

I'm holding out for AI contact lenses.

Combined with compact objects in JDK 25+ this could be a huge breakthrough in scaling JRuby apps. What if your JRuby app's working set could be half as big on the heap, with improved throughput due to smaller objects and less allocator overhead?

A first shot at drastically reducing object size in JRuby, in this case by using more compact wrappers for small Integers: buff.ly/KxZEKBh Other fixes will reduce the size of "immediate" types (numerics, etc) by up to 16 bytes and the size of ALL objects by 5-6 bytes. 🤯

Am I "old" for wanting to discuss a bug and its potential fixes *before* opening a pull request? I'm frequently chided for talking about the issue first rather than just making a PR right away (i.e. "wasting our time"). Personally I tend to reject nontrivial PRs that had no discussion.

Asked ChatGPT to find the best way to choose the smallest Number box for a long value. Several options were discussed with JMH benchmarks. Code looked good, but unfortunately most options introduced tricky negative boundary bugs, despite my requirements. How can I ever trust it to write code?

TFW AI generates *completely wrong* code that almost looks correct. Pretty hard to have any confidence in LLMs for code generation when they creates nonsense like this. buff.ly/kMsOHX0

This new AI world continues to frustrate the hell out of me. I'm seeing pull requests that clearly don't solve the problem, conversations that are automated nonsense, and fake engagement across all social media. No wonder people have no excitement about OSS work lately. ☹️

I get asked about this a lot, so... Would anybody be interested enough in a native-compiled Ruby implementation to fund a fork of JRuby that can be compiled with GraalVM Native Image? It's totally doable, but the work involved is nontrivial and incompatible with the JVM version.

Good progress discussing a Ruby "deep_freeze" feature at this month's ruby-core dev meeting! matz is not opposed to adding the feature, but we need to formalize the protocol and decide how much control user-defined objects should have over the process. We'll make it happen!

Java could do a lot to eliminate its "bloated" reputation by providing space-efficient collections out of the box. HashMap is enormous compared to a direct-addressing implementation and there's no Set impl that doesn't just wrap a Map instance. So frustrating.

Aurora borealis over my home in Minnesota last night. Quite a light show!

Fun fact: JRuby is the only alternative Ruby implementation to attempt Windows support since Microsoft killed IronRuby, and the only one ever to run Rails. We test generating and running a full stack Rails app on Windows before each release. Windows is a PITA to support. 🤬

The apparent decline in Ruby use in recent years has more to do with the fact that people aren't building as much new stuff than anything about Ruby itself. A malaise has fallen on software dev: few exciting projects, skilled devs out of work. Good time to start something?

Looks like there will be no Free Java room at FOSDEM this year (first time in almost 20 years!) so it seems my late January to early February travel schedule has opened up. Anyone have an event they'd like to see me at, talking JVM features and internals or Ruby optimization?

🚨 Time Sensitive 🚨 #Java peers...the 2026 #JavaOne call for papers closes soon. Get your proposals in by the Nov 11 deadline ➡️ social.ora.cl/6017A7UwL

It appears there’s momentum behind a non-forking Solid Queue option in the PR below, but we need help pushing it over the finish line! Even on CRuby, a threaded version can be useful when forked processes are just too heavy. github.com/rails/solid_...

The Solid Queue database-based backend for ActiveJob currently requires `fork` to work, solely because that's the only way to make sure it runs in parallel on CRuby. I want it to work on JRuby, which doesn't support fork, but can run threads in parallel very easily. Help? buff.ly/18ZW69U

What do you call it when you don't ever feel like you really get "in the zone" while coding until you've been at it for 8 hours? It's like I can't really focus until that point. My best days (like today) are the ones where I put in 12-14 hours in a single stretch.

I'm making another attempt to get a `deep_freeze` method added to Ruby. Core folks have rejected proposals for years, until Ractor came along and we got basically the same feature in `make_shareable`. But deep freezing should be its own feature. bugs.ruby-lang.org/issues/21665

I figured out why I don't like Ruby's new Ractor concept: it forces users to work around CRuby deficiencies. To parallelize code with Ractor, users must make all passed object graphs "shareable" even if those objects would otherwise be safe to parallelize with Threads. It feels un-Ruby to me.

It turns out the real Ractors were the Threads we made along the way

Here's one for someone who wants to contribute to JRuby: reimplement UNIXSocket using the JDK support for UNIX sockets: github.com/jruby/jruby/... We currently use jnr-unixsocket, part of our native FFI backend, but it has never worked perfectly.

I'm pleasantly surprised at how easy it has been for my kids to funnel allowance into investments using GreenLight. Interesting educational conversations about stocks and funds and what makes a good investment. Here's some code that gets you $30 back: share.greenlight.com/91792639

I expected at least a few patches to get JRuby 9.4 green on Java 25, but I was wrong. JRuby 9.4 officially supports Java 25 already, I guess! github.com/jruby/jruby/...

New post: Warbled Sidekiq: Zero-install Executable for JVM In my previous post, I showed how to use Warbler to package a simple tool as an executable jar. This post will demonstrate how to “warble” a larger project: the Sidekiq background job server! buff.ly/ZEasNiu

I just used JRuby and Warbler to package up a well-known Ruby-based background job server as an executable jar and it "just works". Single-file deployment on any platform with a JVM and scaling far better than you can get with regular CRuby. JRuby is the future of enterprise Ruby.

That's probably enough of Ractor working for an evening project. I get the idea of the API but the internal details (deep copying and ownership) are going to be really hard for any Ruby to implement efficiently. You could just use threads and frozen objects on JRuby and it would be fast right now. 🤷

New post: "Packaging Ruby Apps with Warbler: Executable JAR Files" Warbler is the JRuby ecosystem’s tool for packaging up Ruby apps with all dependencies in a single deployable file. We’ve just released an update, so let’s use Warbler to create all-in-one packaged Ruby apps! buff.ly/ZdXcCop

I need to run this again to see what might have changed but there is a guide from earlier that I wrote: notepad.onghu.com/2021/jruby-w... But I am delighted that some of the niggling issues might have now been resolved. #JRuby #Ruby #Programming

Turn your Ruby gem into an all-in-one executable jar file with JRuby: * gem install warbler * warble * java -jar mygem.jar ... I forgot how easy this is. Blog post coming. gist.github.com/headius/6739...

Ruby on the re-launched Pebble watch? Let's make it happen @ericmigi.com! buff.ly/na6SE6O

I'm still waiting for my preorder Pebble Time 2 watch, but meanwhile I'm jealous that the only SDK support they'll ship is for Python and JS. Let's change that! mruby would be a perfect way to add Ruby to the PebbleOS SDK. Help me make it happen!

New post: Updating Deprecations with Version Information A short Ruby script and justification for updating all of JRuby's old `@Deprecated` annotations with "since" versions. Easy to run yourself with JBang. Hope it's useful to someone else out there! blog.headius.com/2025/10/upda...

I posted this response twenty days ago at the beginning of the RubyGems drama. Timeline isn't clear but at least some of these possible theories seem to have been in play. Almost like people should let facts come out first. #RubyGems #security #lawyers www.reddit.com/r/ruby/s/sC8...

Ok, someone must have implemented Ruby's "Ractor" API in pure-Ruby on top of threads, right? I want to see what that looks like running with true parallelism on JRuby. There's little demand for Ractors from JRuby folks because they already have shared-memory concurrency with few restrictions.

Any reason why I shouldn't globally replace all bare "@Deprecated" annotations with "@Deprecated(since = 'first release after deprecation')"? Because I really want to do it. It won't be as easy to blame these edited lines, but now the release is right there.

Spring Boot restful app in a one-line command? Love that Ruby magic sometimes... jruby -Ilib -rboing -e 'Boing::Rest.run("HelloApp") do; get_string def hello; "Hello from Boing!"; end; end'

This is the Ruby/LLM future you want. Open source models, FFI instead of C extensions... just write Ruby (or JRuby). rllama: Ruby FFI bindings for llama.cpp to run open-source LLMs such as GPT-OSS, Qwen 3, Gemma 3, and Llama 3 locally with Ruby code. github.com/docusealco/r...

GenAI is a ponzi scheme. Larger and larger investments are necessary to fund new models with diminishing returns and higher costs to run. Reliability is approaching a mathematical maximum. AI companies keep raising prices and still bleeding cash. It's all teetering on the edge.

Whenever I'm forced to use Windows again, I'm shocked both by how slow everything feels, and by how many useless progress bars and "this won't take long" messages I get while installing or updating software.

JRuby 10 officially supports Java 25 and it's green in CI! Of course, we've been testing with 25 locally for months and this required zero changes... but now it's official. 🏆

I'm preparing to launch a new and exciting Ruby gem for JRuby and thought of the perfect name. So I went to rubygems.org to check if it had been claimed. Bad news: it had. Good news: it was a one-off experiment of mine that nobody ever used. I can just repurpose it!