Somewhere in the last couple of years, a strange feeling crept in, and I don't think I'm the only one who's felt it.
It's the sense that I'm not really supposed to enjoy this anymore. That admitting, out loud, in 2026, that I love programming — that I'll happily lose a Saturday to a problem a machine could solve in four seconds — is a little naive. Like I missed the memo. Like there's a version of me who's moved on, who only touches code when it's billable and optimal and defensible, and who looks at the part of me that still finds this fun with a kind of pity.
I couldn't quite name that pressure until I read a short post this week by Chris Kiehl, called "It's Still OK to Enjoy Programming." It's worth your five minutes. He says a simple thing that landed harder than it had any right to: it's OK to enjoy things, and doing so requires no economic justification.
That sentence unlocked something I'd been circling for a while, so I want to build on it — because I think the pressure he's describing is real, I think I understand where it came from, and I think it's quietly taken something from a lot of us that's worth taking back.
The doom stream doesn't argue with you. It soaks you.
Here's the thing about the "programming is over" narrative: it rarely wins by convincing you. It wins by repetition.
Open any developer feed right now and you'll find a steady drip of it. AI is going to replace you. The craft is dead. Why would anyone learn this now. The junior path is gone. It's all just prompting. Every scroll, another small dose. And none of it argues with you directly — it doesn't have to. It just soaks into you, a little at a time, until one day you sit down to build something for the pure pleasure of it and a quiet voice you didn't install asks: what's the point? A machine could do this faster. Why bother?
Kiehl put his finger on exactly this — that the more of the doom stream you consume, the more it infects you, and that it disappears the moment you disconnect. I think that's precisely right, and it rhymes with something I keep noticing about this era more broadly: we keep letting an ambient feed rewrite how we feel about things we used to love, not through any single convincing argument, but through sheer saturation. You don't decide the joy is gone. It just gets slowly crowded out by a mood you absorbed from a thousand posts you barely remember reading.
And underneath that mood is a premise so quiet you might never notice you swallowed it. It's worth dragging into the light, because once you see it, it loses most of its power.
The lie hiding inside the despair
The doom stream smuggles in one assumption, and it's this: that the value of doing something is its economic value.
That if a thing can be done more cheaply by a machine, then doing it yourself is pointless. That your joy in the doing is a kind of inefficiency — a sentimental attachment to a task that's been rightfully automated. That worth and worth money are the same thing.
Say it plainly and it's obviously false. But we almost never say it plainly, which is exactly why it works on us.
Because think about how absurd it is applied anywhere else. Nobody tells you your evening run is pointless because a car is faster. Nobody says you shouldn't cook dinner because restaurants exist and are better at it. Nobody looks at someone playing guitar badly in their bedroom and says, "You know Spotify has professionals, right? What's the economic case for this?" We'd find that insane. We understand, instinctively, that people do things because the doing is the point — that the value was never in the output, it was in the experience of making it, the small private satisfaction of having done a thing with your own hands and mind.
We understand that about everything except, apparently, the thing we spend most of our lives doing. Somewhere along the way we let the market convince us that this particular love — unlike running, cooking, music, gardening, painting, every other human thing — has to earn its keep. Has to be defensible. Has to show up on an invoice or it doesn't count.
It doesn't. It never did.
The best thing I built this year was worthless
I'll be concrete, because this isn't abstract for me.
The things I've built recently that I'm proudest of — that I think about, that made me feel most like myself — are almost all completely worthless by any economic measure. A tiny tool nobody but me will ever use. A problem I solved the slow, hard way on purpose, long after I knew the shortcut, just to feel it click. Things I could have prompted into existence in a fraction of the time, that I chose to struggle through by hand because the struggling was the entire reward.
Kiehl mentioned building a buggy Breakout clone with terrible collision detection, by hand, because he enjoys it — and noting that it doesn't bother him that an AI could one-shot it, "because that's not the point." I felt that in my chest. The point was never the Breakout clone. A perfect, instantly-generated Breakout clone would have given him nothing, because the thing he was actually after wasn't a game — it was the hours of being absorbed in a problem, the small victories, the specific texture of figuring something out yourself. The output was disposable. The experience was the whole thing.
That's the part the economic lens can't see, and never could. It measures the artifact and misses the aliveness. It looks at the bad Breakout clone and sees a worse version of something a machine makes better. It cannot see the person who was happy while making it, because happiness doesn't have a line item.
The quiet rebellion
So here's where I've landed, and it feels almost defiant to say it, which tells you how strange things have gotten.
The most radical thing you can do as a developer right now is not to resist AI, or out-hustle it, or have the correct take about it. It's to still find this beautiful. To sit down on a weekend and struggle, on purpose, with a problem a machine could solve in seconds — and to enjoy the struggle, precisely because no one asked you to and nothing depends on it. To build something gloriously useless. To learn a thing that will never once come up at work, just because understanding it is its own reward.
This isn't anti-AI. I use the tools; they're genuinely wonderful, and some of the fun I have now is because of them. It's not nostalgia, and it's not pretending the economics of our jobs haven't shifted — at work, where you're paid to produce value, they obviously have, and that's fine. That's work. Work was never the whole of it.
It's just a refusal to let the market price the one thing it was never entitled to: your wonder. Your right to do a hard thing badly, by hand, for no reason at all, and to love every minute of it.
It's still OK to love this
So if you've been feeling that quiet pressure — that you're somehow behind, or naive, or sentimental for still enjoying this — I want to say the thing Kiehl said to me, and pass it on:
It's still OK to love this. Not because it's useful. Not because it'll pay off. Not because you can justify it to anyone, including yourself. It's OK to love it for the same reason anyone loves anything — because the doing lights something up in you, and that was always, by itself, enough.
The doom stream disappears the second you close the tab and open the editor for no reason at all. Build the useless thing. Struggle with the solved problem. Lose the Saturday. None of it has to be worth money.
It just has to be worth doing — and only you get to decide that.
With real thanks to Chris Kiehl, whose post "It's Still OK to Enjoy Programming" is the reason this one exists — go read his, it's shorter and kinder than mine.
And a question I'd genuinely love answered in the comments: what's the most economically worthless thing you've built or learned recently, purely because you wanted to? I'll start — I spent a weekend reimplementing something that already exists, works better, and ships with my OS, entirely by hand, for no reason at all. Best time I've had in months.
