This morning, I noticed I have a blog here, and I last wrote anything in 2021. I tried to think through what has changed since then across the various areas of software development, and I realised that so much has changed that I couldn’t even begin to list it all.
I’ve been involved with this wonderful thing we call programming—and design as well—for quite a long time, since sometime in the late eighties. But honestly, that hardly means anything, because everything is constantly changing, and not just the technologies but the fundamental principles as well.
There is always something to learn, and you can learn just as much from younger generations as from your own experience and mistakes. I’ve always enjoyed working with younger people, because their enthusiasm often helps carry you through difficulties that would otherwise be hard to overcome—and also through those things that one might otherwise get bored with, do on autopilot. And maybe this is what senior developers forget so often: learning from younger people.
And by the way, this is true in other areas of life as well. Innovation is always very important, but it’s hard to stumble upon truly innovative ideas, because it’s easy to get lost in the maze of hype and the background noise created by reinventing the wheel a thousand times. What’s funniest is that even today, I still work manually—without LLM tools—at my workplace, because my employer hasn’t approved their use yet. Banking sector. Lots of approvals. Slow process. Of course, I do use AI in my personal projects—they’re clever, useful, and a little bit scary.
As for the future… I’m fairly sure that I won’t retire as a software developer, quite simply because there won’t be a need for me. People often deny reality because they’re afraid. This is one of those phases in the industry right now. Many are afraid, many are enthusiastic, many force pragmatism upon themselves—I think all of these are valid responses, but we’ll see what happens.
In about three years, I’ll read what I wrote again and reflect on it :) At the moment, I believe that by then almost all software will be generated automatically, and that LLMs will soon appear that transform programming languages themselves and generate binary code. It’s even likely that AI will be embedded directly into hardware. This entire industry that we currently call software development will change so radically that it’s hard to even imagine today.
In any case, I’m going to start learning something completely different, go back to university, and after thirty years of programming and software design, it will be good to dive deeply into something else. In the meantime, I’ll keep working—of course—and if I turn out to be wrong and I’m still needed in three years, then I’ll have a good laugh at myself.
Top comments (0)