The thing that brought me joy
66 points
8 hours ago
| 15 comments
| stephenlewis.me
| HN
overgard
18 minutes ago
[-]
I think fundamentals are more important than ever. How can you debug without understanding it? "Claude please rewrite" is not good engineering.
reply
frizlab
5 hours ago
[-]
I love this, it resonates so deeply with me. Code is, for me, joy. I spent a little more than an afternoon writing a parser to parse a new ad-hoc file format I created to represent the IDs (class name and ID names) I will use in my CSS, and it was just fun. Sure some AI could probably have written that for me, but for what? So I can dig directly back into complicated actual engineering issues? Where would my breaks be?
reply
abeindoria
4 hours ago
[-]
Ironically I have a somewhat of a different view - I love rubber ducking and tinkering with LLMs. Sometimes they come up with a use case that I would not have thought of, but I would have liked to have maybe 2 weeks later. Other times it is nitpicking each others' code etc.
reply
ainiriand
5 hours ago
[-]
I think we are presented with a false dichotomy here, as you can use llm tools for menial tasks and code whatever scratches your itch at the same time. For me, I really do not enjoy writing any frontend, html, javascript, whatever; I just want to bring some website I need to light. I focus on other code and that is what brings me joy.
reply
altmanaltman
24 minutes ago
[-]
I mean that's perfectly valid for a hobby project, but I think the argument is from the pov of a company seeing your time as a resource. In that context, it is obvious why it makes economic sense to spend your time on the actual, complex issues, assuming AI can handle the basic tasks. My job doesn't give a f about me finding joy, I can raw dog all the code I want in my free time/for hobby projects.
reply
exitb
4 hours ago
[-]
> Sure some AI could probably have written that for me, but for what?

One reason would be to raise the ceiling of what your project can do within the budget of time and motivation you have. Or, as it often happens, to be able to finish the project at all.

reply
lordnacho
2 hours ago
[-]
This is artisans vs industry.

You can get a hand-crafted, beautiful, solid, chair made by a guy who knows all the tricks of carpentry. He's spent his whole life perfecting the technique of how to drill the holes, how to fit everything together, how to balance the chair.

It used to be the only way.

Then someone invents a machine that can make you a hundred crappy chairs. Sometimes the legs don't fit in the seat, sometimes the chair isn't balanced. But it's close enough often enough.

On top of it, the new tool is not in the lineage of the old tools. It doesn't FEEL like you are crafting a chair.

reply
conqrr
57 seconds ago
[-]
Except that chairs (or anything that's automated today, even cars) don't rapidly evolve or are dynamic like Software is.
reply
weiliddat
4 hours ago
[-]
I feel similarly. Sounds a bit like other crafts that were later industrialized and (partially) mechanized, like woodworking and carpentry.

One can certainly enjoy the laborious handcrafted process of building your own table, and yet go back to a shop that churns out cheap furniture that’s nonetheless useful for many others, and see the value in both.

Obviously there’s more degrees of freedom in software, but I’m trying to see it that way to rationalize how I’m feeling with the current state of things.

reply
senko
5 hours ago
[-]
> It’s not that the agents are now producing flawless code. I spent a good 20 minutes yesterday watching one tie itself in knots trying to write a regex: first in Sed, then in Bash, and finally in Python (six times).

This sounds very strange.

I'm using Claude (Opus 4.5 via Code) every day and it's very good with regexes, sed, awk and similar bash oneliners.

We don't know what the author asked it to do, but this smells like the problem started at least several messages before that.

To author's point: code brings me joy. I'm currently learning Zig, for no reason whatsoever other than intellectual challenge and I, subjectively, like the language. I'm writing silly little programs that nobody will ever see. It's fun.

Then I switch over to a paid project, and claude[0] another task from my backlog.

There's code, and then there's code. You can find joy in some code and absolutely want to avoid coding in something else.

[0] code using Claude

reply
monooso
4 hours ago
[-]
> We don't know what the author asked it to do, but this smells like the problem started at least several messages before that.

Author here. It was what I assumed would be a fairly simple task, fixing some duplicate closing frontmatter delimiters.

I think the LLM took a wrong turn early on, and then just spiralled. It was morbidly fascinating watching it rabbit hole.

reply
senko
4 hours ago
[-]
> It was morbidly fascinating watching it rabbit hole.

Oh yes, it's like slow-motion train wreck.

Once it gets off the rails, the best thing is to nuke the context, reset the state (git reset --hard or equivalent) and retry, this time with foresight.

reply
analogpixel
3 hours ago
[-]
Some people like the making and tinkering with 3D printers, but have no actual projects they want to make with them, nor do they have the skills to use CAD to design them. On the other side are the people that have no interest in 3D printers, and just use them to implement their ideas.

I'm starting to think programming might become the same thing; people that program will be the people that just like to tinker with code and have no real ideas to implement, while the people that have ideas will use LLMs to implement the code so they can just work on their idea.

I've seen the same thing with Linux. There are people that like to tinker with Linux, playing with every setting and every window manager, but don't actually use the computer to DO anything, and then there are people that don't care what the OS is, and are just using the computer as an actual tool to get something done.

reply
heikkilevanto
2 hours ago
[-]
I think there must be a continuous spectrum between the extremes you describe.
reply
card_zero
3 hours ago
[-]
I suspect there's a synergy between the two.
reply
chr15m
1 hour ago
[-]
You don't have to leave the command line. Claude Code, and other agentic tools like it, make you feel like you've left the command line because they capture the whole terminal and leave you with less agency. There are tools like aider (and my own project runprompt) which let you retain more control in your terminal and in my experience feel much less disempowering, while still giving you a productivity boost.
reply
chasil
1 hour ago
[-]
My father passed away a year ago, and I felt a deep well of depression, an inescapable draw that tainted every action, every color of feeling, that sapped my very soul.

I had booked a trip to Cartagena. I hadn't realized that I arrived on Fat Tuesday, and I don't think that it would have mattered.

Arriving there, bathing in the insubstantial "joy d'vivre," the sinking feeling lifted.

I remember that well, and it's cure. I love Latin America for this reason.

reply
jstummbillig
5 hours ago
[-]
If it truly brings you joy we have hat covered: it's a simple enough hobby!

The actual issue is that then you need something still that makes money. I think, for a programmer, that's fairly unproblematic too, for the foreseeable future: all those agents will need direction. Anyone can do that up to some level of complexity on their own, sure, but it simply is hard for humans to structure requirements and reason about a big enough systems and I don't see demand for those decreasing.

reply
nehal3m
4 hours ago
[-]
I write for fun, to organize and articulate my thoughts, and I love doing that in vim. The same is true for note taking (I just write .md files and sync them with syncthing). I also like working neomutt. It's just fun working with (relatively) simple, stable tools that seem to grow on you over time for every day tasks. Writing code is just one of those things.
reply
PlatoIsADisease
3 hours ago
[-]
Interesting, I suppose I'm still doing debugging and finding solutions to problems even with AI coding. I think my dopamine was from problem solving, not necessarily doing the math/logic. It helps I get paid for solving problems, so more problems solved = more money. And I'd like to emphasize, I still debug.
reply
jwpapi
3 hours ago
[-]
I think it seems to crystalize out a skill set of prompt engineering, debugging, system design (especially safeguard) design and potentially testing to be the most efficient for productivity. It depends obviously what kind of coding you do. The more standard it is, the more AI can help. When you need to engineer new patterns, math, system design logic is more important.
reply
idopmstuff
3 hours ago
[-]
> He’s not wrong, but it does make me wonder: even if the code was not the point, what do you do if it was the thing that brought you joy?

You still write the code! We do lots of things that could be done more efficiently. People build their own furniture and make their own clothes and brew their own beer. If you love doing something for its own sake, keep doing it.

reply
aabajian
3 hours ago
[-]
AI is making us face the reality of non-physical goods and services. If a product exists solely as data (writing, music, code, drawings, movies, etc.) then the value is in the idea and its distribution, and less on its production.

I'll give some examples:

Novel algorithms: PageRank, BitTorrent, Like buttons, disappearing pictures, etc.

Artistic styles: Cubism, impressionism, Wes Anderson, etc.

The above algorithms are (relatively) straightforward to implement and could be implemented by Claude Code in a matter of hours if not minutes. But, you'd still need a means to distribute them.

Similarly, you can have AI generate an image in any of the above styles (or a combination thereof), but the image won't have intrinsic value unless you can finds a means of (profitable) distribution.

Put another way, there's limited value in being able to master physical tasks (playing piano, typing fast), but fundamental skills that lead to creative innovation will remain important...along with being able to package/market/distribute the AI-implementation of your ideas.

reply
card_zero
2 hours ago
[-]
You could imagine beings who live on an abstract plane, creatively playing around with abstract thoughts and producing abstract goods. Except you couldn't imagine it in any detail, because it has no details, because its abstract. It's easier to imagine that they live in a virtual world, and play creatively with virtual objects in the course of producing virtual goods that owe a lot to real physical goods. But that's tantamount to learning physical skills. Besides, for the time being, the world we inhabit is an amalgam of the virtual with the physical (not to mention the abstract). I'm saying that if you lock yourself in a box and think abstract thoughts you won't innovate much. It worked for Descartes with his oven, but that's unusual. Usually we need to play with a somewhat physical environment that's friendly to our physical monkey-shaped bodies and senses, in order to be creative, so physical skills aren't going away in the foreseeable.
reply
jrm4
5 hours ago
[-]
AI assisted coding (and broader idea exploration) is 100% bringing me joy in this way.

I don't code for a living in any way, but I teach IT. And for years and years I've had little script ideas and tasks (e.g. music organization) that worked decently, but also life got in the way and I have that thing where I want it to work just right etc, and now that the pipeline is orders of magnitude shorter, man this stuff is FUN for me again.

reply
Barrin92
4 hours ago
[-]
>Good enough that choosing to focus on the fundamentals seems rather foolish.[...]but most of the time it sticks the landing. At this point, learning Awk seems about as sane as mastering the loom

The joy aspect aside, if you're writing code professionally obviously that is no substitute for knowing how Awk works because you still need to make sure that what it produced is correct, because I sincerely hope that you're not pushing code into production, in particular not a piece of code some tool got wrong for 20 minutes, without understanding what it does.

If that were an intern rather than a machine and you as the programmer don't know how the code works you're not doing your job. AI systems have not freed people from acquiring "arcane knowledge" it makes that knowledge more important than ever, the only thing it does is type faster than you do.

reply
monooso
3 hours ago
[-]
Author here...

> I sincerely hope that you're not pushing code into production... without understanding what it does.

I do not. That hasn't changed.

> ...the only thing it does is type faster than you do.

LLMs can do many, many things that I cannot. Their breadth of knowledge dwarfs mine.

reply