Skip to main content

Seriously uncool Java?

A couple of podcasts I have listened to recently have flamed Java. It is a good "kusitolppa", a seriously uncool technology.  I'm pleased about not having to write Java anymore. Yet, I'll risk career suicide and take a stance to defend it against some of the low blows. After all, it was my bread and butter for quite many years.

I get it; writing Java is not in fashion. It is likely a wrong choice for a modern startup. The language is full of mysterious cargo-cultish traditions, is verbose, has annotations on top of annotations and you have to use an editor which eats half of your memory (the other half is used by the Java process you are working on). Traditional Java and JDKs are ill-fitted for short-running processes like lambdas; it needs time to warm to get up to speed and needs more than one core to work efficiently. It has horrible dependency management tools by today's standards. Running a JVM process requires all kinds of weird flags, its defaults are unlikely suitable for any given project.

An often pointed out defect in the old-school Java frameworks is their excessive use of threads. Granted, the traditional, solid but inefficient, thread-per-request model has been pervasive for too long. It's funny why, since the standard library has had the fork-join work-stealing type thread pool in for a decade. You are supposed to use that, preferably through a library or a framework, and never create threads yourself.

Speaking of frameworks, It's not trivial to figure out what is a good JVM framework. The de facto choice is Spring Boot or something expensive from RedHat. I would steer clear of those enterprisey frameworks and pick something like Micronaut or Quarkus. They have a fast startup speed,  don't need massive thread pools (non-blocking IO), and use code generation rather than runtime reflection. I won't even try to do or link any study on their performance compared to other frameworks from other languages since it is notoriously tricky, but I'm sure they are in the triple-A class.

GraalVM's ahead-of-time compilation can speed up startup and reduce the docker image sizes. The new CG's, ZGC and Shenandoah, have finally enabled massive heaps with low latency stop-the-world pauses. Since JDK 8, tiered JIT allows complex, heterogenous codebases to run fast after a warmup period. If a warmup time is not an option, Quarkus with Graal AOT compiled native image should do the trick.

I could go on about how slow and somehow uncomfortable Jest feels compared to plain old JUnit, and what a mess I see when I want to check the TypeScript definitions from the editor compared to the dull but very readable JavaDoc. I love writing TypeScript in VSCode, but I do spend a lot of time resolving all kinds of weird double import errors. IDEA was laggy and slow to start, but it did a lot of coding for me.

My conclusion here is that I would not personally start a new Java project and certainly would not build a startup with Java, but at the same time, I should not discard the JVM ecosystem so blatantly. If I were to choose the JVM, It's a no-brainer to use something like Kotlin or Clojure to get the best of both worlds, a solid runtime, and an expressive programming language.

Labels:

Comments

Post a Comment

Popular posts from this blog

I'm not a passionate developer

A family friend of mine is an airlane pilot. A dream job for most, right? As a child, I certainly thought so. Now that I can have grown-up talks with him, I have discovered a more accurate description of his profession. He says that the truth about the job is that it is boring. To me, that is not that surprising. Airplanes are cool and all, but when you are in the middle of the Atlantic sitting next to the colleague you have been talking to past five years, how stimulating can that be? When he says the job is boring, it is not a bad kind of boring. It is a very specific boring. The "boring" you would want as a passenger. Uneventful.  Yet, he loves his job. According to him, an experienced pilot is most pleased when each and every tiny thing in the flight plan - goes according to plan. Passengers in the cabin of an expert pilot sit in the comfort of not even noticing who is flying. As someone employed in a field where being boring is not exactly in high demand, this sounds pro...
Post a Comment
Read more

PydanticAI + evals + LiteLLM pipeline

I gave a tech talk at a Python meetup titled "Overengineering an LLM pipeline". It's based on my experiences of building production-grade stuff with LLMs I'm not sure how overengineered it actually turned out. Experimental would be a better term as it is using PydanticAI graphs library, which is in its very early stages as of writing this, although arguably already better than some of the pipeline libraries. Anyway, here is a link to it. It is a CLI poker app where you play one hand against an LLM. The LLM (theoretically) gets better with a self-correcting mechanism based on the evaluation score from another LLM. It uses the annotated past games as an additional context to potentially improve its decision-making. https://github.com/juho-y/archipylago-poker
Post a Comment
Read more

Careful with externalTrafficPolicy

On a project I am working on is hosted in an EKS cluster with the NGINX ingress controller (the one maintained by Kubernetes). It is deployed using it's official official Helm chart, which I realized, after a lengthy debugging session, was a mistake. The initial setup I aimed to improve had several flaws. Firstly, we were using the AWS Classic Load Balancer in front of the nginx ingress in the cluster, which has been deprecated for some time (years?). Continuing to use it makes little sense to us. The second issue was that we were only running one(!) nginx pod, which is quite sketchy since the exposed web services had essentially no high availability.  I switched to the Network Load Balancer (NLB), which was straightforward - I just needed to change the ingress-nginx service annotation to specify the load balancer type as NLB: service.beta.kubernetes.io/aws-load-balancer-type: nlb However, increasing the replica count turned out to be tricky. When I bumped it up to two, I began to ...
Post a Comment
Read more