You’ve probably felt this. Someone mentions a technology — Cloudflare Tunnel, RAG, Termux, vector search — and a small door closes in your head. “That sounds powerful,” you think, “but it’s not for me. I’m not an engineer.”
For most of my life, that feeling was right. Knowing the name of a technology got you almost nowhere. Between the name and a working result there were months of study, and a level of skill I was never going to reach.
That gap has gotten much smaller, and it has changed how I learn. Here’s how I think about it now. It’s not a rule for everyone — just what has worked for me.
These days I don’t try to master a technology before I use it. I try to know it exists. The hard part — the syntax, the bugs, the wiring-up — is more and more the part an AI assistant handles. What I bring is mostly knowing what to reach for.
I’m a sixty-year-old with no coding background, and I’m a fair example of this. In my last post I built my first app in a day. I could do that not because I understood the technology, but because I’d heard the names of the parts I needed. Let me explain how that even works.
My One Source
For anything to do with computers, networks, or AI, I have basically one source. Not the internet, not courses, not workmates. My older brother.
He’s the real thing — an independent IT and AI professional, genuinely good, the kind of person who builds things most people only read about. And he can’t help himself. When some new technology excites him, he calls to explain it. The trouble is, he can’t keep it short. The calls run long, and longer, until we both give up.
So the real teaching happens somewhere else. A few times a year, our family gathers at the farm by the Boseong River, in the Korean countryside, where my mother lives. (If you’ve read the origin of all this, you know that farm.) There, over meals and long afternoons, my brother finally has the time to show me what he was trying to explain on the phone.
The Notebook and the Year-Long Delay
Here’s the part that used to look like a weakness, and now looks useful.
When my brother explains a new technology at the farm, I don’t really understand it. Not deeply. I’m not at the point where I could use it, and I know it. So I do the only sensible thing. I write down the name and one line about what it’s for. Then I go back to my life.
Maybe a year later — sometimes more — I run into a problem where that name fits. I dig out the note. Ah. This is the thing for that.
I first heard “Nginx” and “Cloudflare” from my brother more than a year before I had any use for them. They sat in a notebook, meaningless, until my wife’s messy folder of audio gave them a job. Then I pulled them off the shelf.
In the old days, that year-long delay would have killed it. You’d forget the manual and give up. Now the delay costs almost nothing, because I never needed to keep the technology in my head. I only needed to remember it existed.
What Changed
Think about where the hard part used to be.
The thing that kept people like me out was rarely knowing that something existed. The encyclopedias have always been full of names. The hard part was building it — turning a name into a working thing on your own machine. That’s the wall I stood behind for sixty years.
AI lowered that wall a lot. When I built my wife’s player, I didn’t write the code that connects Nginx to the app, or work out the exact Cloudflare settings. I described what I wanted in plain words, and the AI did the part that used to need an engineer.
So the value of things has quietly flipped, at least in my experience:
- What used to be cheap — knowing a name — is now the part that matters most. It’s the seed. Without it, you don’t even know what to ask for.
- What used to be expensive — building skill — is now something you can rent a sentence at a time.
That’s why my brother, and the farm, feel more valuable to me than ever. Someone who just tells me “this exists, and here’s roughly what it’s for” is handing me the one thing I can’t get any other way. And a place to sit with him a few times a year keeps it coming. The farm by the river isn’t only where my family gathers. It’s where I collect names.
Why I Finally Opened Claude Code
For a long time, “Claude Code” was just another name in my notebook. My brother had mentioned it many times — you should really try this one — and I kept meaning to. I’d even dabbled at work once, building a document-search tool with a different AI assistant and a method called RAG. So I wasn’t a complete stranger. I just hadn’t started using it for myself.
My wife’s audio is what finally pushed me. I wanted to build her something real, and I wanted to do it then. So I opened Claude Code, described the app, and pointed it at an example to follow — an audio app my brother had already built and put online. Make me something like this, I said, but for her recordings.
A day later, the core of it worked.
I want to be careful here, because I don’t like hype. It wasn’t perfect in a day. The search and the small fixes took weeks of daily use. But the jump from nothing to a working app my wife could open happened in an afternoon — done by someone who can’t code, with little more than the names of the parts.
What This Might Mean for You
If you’ve spent your life thinking technology is for other people — younger people, technical people — it might be worth a second look. The rule I grew up with has changed.
You don’t have to study a technology before you’re allowed to use it. You can do something easier:
- Collect names. When you hear something exists — a tool, a method, a service — write it down with one line about what it’s for. You don’t need to understand it. You’re building an index, not a manual.
- Wait for a problem. Don’t force it. One day a real frustration in your life will match a name on your list.
- Let the AI close the gap. Describe what you want in plain words, point it at an example if you have one, and let it handle the part that used to need an engineer.
Find your source, too — the person, channel, or community that keeps feeding you names. Mine happens to be my brother and a farm by the river. Yours will be different. But that source may be the most useful thing you can have right now, because it gives you the one thing no machine hands you: knowing what to reach for in the first place.
Where This Goes Next
Everything in this series is a chain of names pulled off that shelf. Termux — heard at the farm one winter, used months later to turn a cheap tablet into a server. ElevenLabs — heard only a few months ago, used to transcribe hundreds of hours of audio. Each sat in my notebook, meaningless, until a real need gave it a job.
The next post is where that starts in earnest: how a cheap Android tablet — the kind you’d hand a child — became a Linux server quiet enough to run in an eighty-year-old man’s living room.
If you’re following along, please subscribe.
Key Takeaways
- These days, knowing a technology exists matters more than knowing how to build it.
- Building it — the old wall that kept non-engineers out — is more and more the part AI handles.
- You don’t have to keep technologies in your head. Keep a notebook of names and one-line purposes, and pull them off the shelf when a real problem shows up.
- A mentor or source who feeds you names may be the most useful thing you can have.
- I built my first app in a day not by understanding the parts, but by knowing they existed and pointing AI at an example.