Bill MacKenty

Automation Has Made Programming Less Fun.

I still remember the flicker of excitement I felt when I wrote my first line of code at 14...

The screen blinked back at me, a silent acknowledgment of the journey I was embarking upon. Over the past 41 years, coding has been more than just a profession; it's been a passion fueled by curiosity, creativity, and the thrill of solving complex problems. But recently, I had an experience that made me question the evolving landscape of programming and its impact on the joy it once brought me.

A few weeks ago, I decided to create a simple ASCII-art 4X space game—a project that, in the past, would have been a delightful challenge filled with hours of brainstorming, debugging, and incremental victories. This time, however, I turned to a Large Language Model (LLM) to assist me. In just about two hours, the game was complete. No hurdles, no late-night problem-solving sessions, no trial-and-error. And yet, instead of feeling accomplished, I was... bored.

Coding has always been akin to solving a intricate puzzle. Each bug fixed and each function optimized brings a sense of achievement that's hard to replicate. The process demands patience, logical thinking, and creativity. It's not just about the end product; it's about the journey—the countless trials and errors that lead to the final result.

When an LLM can generate code in a fraction of the time, it strips away the challenges that make coding rewarding. The automation of problem-solving turns an engaging process into a mechanical one. The excitement of unraveling a complex issue diminishes when the solution is handed to you on a silver platter.

As an educator, I see the same patterns emerging among my students who are learning to code. The allure of quick solutions is tempting, but it deprives them of the fundamental experiences that build proficiency and confidence. Struggling with code isn't a setback; it's a crucial part of the learning curve. It's through debugging and iterative problem-solving that students develop a deeper understanding of programming concepts.

When students rely too heavily on AI-generated code, they miss out on the opportunity to think critically and develop their problem-solving skills. The "eureka" moments that come after hours of hard work are invaluable. They not only reinforce learning but also build resilience and a growth mindset.

This isn't to say that LLMs and AI tools have no place in programming—they undoubtedly increase efficiency and can handle repetitive tasks with ease. However, it's essential to strike a balance. For seasoned programmers like myself, perhaps it's about using these tools to handle mundane aspects while reserving the more challenging problems for manual coding. For students, it might mean using AI as a learning aid rather than a crutch.

Coding is changing rapidly with the advent of AI and automation. While these tools offer incredible benefits, they also pose questions about the future of programming as a fulfilling craft. For those of us who find joy in the challenges of coding, it's important to remember why we started in the first place. And for the new generation of coders, embracing the hard work and the hurdles isn't just beneficial—it's essential.

The next time you sit down to code, consider taking the longer path. Embrace the difficulties, relish in the trial-and-error, and remember that sometimes, the struggle is where the real fun lies.