Language acquisition requires a significant amount of congitive effort. After moving to the UK, I realised that as an ESL (English as a Second Language) learner, I had to put in more effort to improve my English speaking skills, since communiating with native speakers is completely different. As someone who loves travelling to Japan, watching Japanese TV programmes, reading manga and playing on the Switch, over time, I procrastinated a lot when it came to committing to learning Japanese.

But today, I’m not writing about learning English or Japanese – I’m writing about learning Go, a programming language.

Pain in Java

I first used Java in 2010 while attending a course at university. I’ve been using Java for backend development since 2013 because of my work. Since then, I haven’t changed my tool at all. In the meantime, we faced many challenges.

Java has abstractions. I don’t oppose abstraction – it does make the code more maintainable, but only if it’s implemented correctly. From time to time, “perfectionists” or learners tried to abstract as much as they can. They eventually created abstraction hell, making the code difficult to understand and debug.

“Why is there an abstraction?”
“It’s because we want to adopt the Strategy Pattern.”
“So, how many alternative behaviours are there?”
“Umm… One, possibly three.”

Java is a bloated programming language. I was using Java 6 in 2013, moved to Java 8 in 2017, and switched to Java 11 – all the way to Java 21 when I was leading a product team. Last November, I built Mahjong101 with Java 23. There were tons of changes in Java over these 11 years. I can still remember the first time we learned Stream and Optional in 2017. I spent plenty of time learning the mechanics and suggesting others try the new libraries. Last year, when I was visiting a development team, they asked me for coding advice. After listening to my advice, they didn’t know what I meant. I realised they were still using Java 6. Although we were using the same language, it felt like we weren’t!

Apart from Java, its well-known framework, Spring Boot, is a nightmare. Yes, it’s a Swiss Army knife – there are lots of tools. With proper configuration, setting up a database communication, RESTful APIs and security should be simple. However, it only works smoothly if you’re highly experienced and the configuration is perfectly written. If something goes wrong, unforutunately, welcome to the hell of IllegalStateException, where you’ll get hardly any clues about what’s causing the issue.

Furthermore, Spring Boot is even more bloated than Java. Many libraries under Spring Boot contain layers upon layers of abstraction. First, and again, this makes it difficult to understand and debug. Second, it includes too many unnecessary dependencies. Over time, vulnerability reports required us to upgrade certain libraries. Sometimes, we had never even heard of the library before, but it had been bundled in with another dependency. Worst of all, some of these upgrades were backward-incompatible. It took us over a month to upgrade Spring Boot across all 50+ microservices from 2.6 to 3.1.

Enough ranting – there are things I love and hate about working with Java. But there was no other option in my company, so I never had intention to try something new.

But now, I’m working on my solo project. Instead of only sticking to the technical stack I know well, I should try something new.

So Why Go?

When it comes to choosing a new programming language, Go is always the first that come to mind.

  • Go is designed for simplicity. I don’t need to include tons of dependencies to implement a RESTful API – everything is included in the standard library.
  • I prefer both static and strict-typed languages, whereas I found working with JavaScript messy.
  • Go’s pointers remind me of my struggles with C++ when I was at university – and I want to overcome them.
  • After 10+ years working in an OOP world, I want to gain some experience in procedural programming.
  • Many companies use Go as their primary, or even exclusive, backend language, including those in the financial industry.

Since so many people praise Go, why not give it a try? Even if I don’t like it, I’ll still broaden my horizons and equip myself with a new tool.

The Plan

In fact, I played around in A Tour of Go last summer, but I got bored and gave up a few days later. Now, I’m taking a different approach.

  • Spend at least 100 days with Go – For consistency.
  • Solve LeetCode problems with Go, regardless of difficultly – To master the basics.
  • Build my next solo project(s) with Go – To apply what I learn and go beyond the basics.

Let’s see what happens in 100 days.

Do you feel the same frustrations with Java? How do you overcome them? What’s the next programming language you want to learn?