I personally used to enjoy the challenge of it. I enjoyed the craftmanship. I am so glad I got to enjoy the challenge of implementing IETF specs and the debugging interop etc... it's so sad to me now that people just bruteforce all this with agents.
I wonder how carpenters felt when powertools started taking off?
I reckon powertools is the wrong comparison, my grandfather was a cabinetmaker(born in the 1930's) and he never shied away from using a powertool & any contractors/tradesman largely raves about their favourite brand of powertools.
I think the more apt comparison would be a carpenter to a mass produced factory. If you've read any of Yegge's recent ramblings than factories of factories seem to be what the AI focused peeps like to produce.
> I reckon powertools is the wrong comparison, my grandfather was a cabinetmaker(born in the 1930's) and he never shied away from using a powertool & any contractors/tradesman largely raves about their favourite brand of powertools.
I'm sure there's some survivorship bias at play here. You don't hear what the tradesmen who shied away from using powertools thought about them -- because they ceased to be tradesmen.
In the same way, in the future, all software engineers will be raving about their favorite coding agent.
He also made the most beautiful hand crafted woodwork I’ve seen.
I dunno, I feel my comparison is more apt. AI tries to anything & everything. A powered drill just drills, and saves your arm/wrist from a manual drill.
A better analogy for an individual might be a cnc lathe. You can make anything you want perfectly but if you enjoyed wood turning then you may not get the same satisfaction from the process of running a cnc.
I think there’s still value in hand made furniture with little to no power tool involvement. There’s certainly value in creating furniture yourself in this way.
It doesn’t give answer the question about the people who loved that job , the many old school carpenters who had to adapt or give up.
I wonder if hand coding will become mostly a hobby and maybe a niche market. I’m not saying I like it, that’s tbd, but I wonder about it.
It's not crazy to feel this way. When the barrier to entry drops dramatically very quickly, it's natural to re-evaluate all your priorities. I personally flip between apathy and excitement.
User expectations are also going to grow. If you're already a dev, you're already ahead if you keep up. It's always been the case that someone could release a free package that makes your personal proprietary software suddenly hard to justify.
I don't think anyone is worried about user expectations. It's management expectations. The output expectation goes up, while the quality of that output goes down, but is still your responsibility.
I'm not surprised professional software developers are not enjoying AI development.
Roughly speaking, you might have taken a task that was in the control, flow, or relaxation zone, closer to the center there, and lowered the difficulty so much that now it's flat out in the middle of the boredom zone.
My answer to this is to either seek out much harder problems (moving it back up into challenge, for example), or to change to a task where I have lower skill and so am in the learning mode rather than the execution mode, which I find much more pleasant and compelling than boring.
You change the goalpost, basically, or give yourself a handicap, to bootstrap yourself back up into motivating regions of the graph. Just a model, but I think it's one that is really useful.
Precisely so. This is also why you find people, mid career, suddenly having that crisis and trying some different things to break up the monotony. If we had a more stable system of employment, things like sabbaticals become really essential to reset the joie de vivre you lose when you start to be an expert executing the same plan over and over.
I don’t think it's the same at all. Nobody is interested in watching computers play against each other, so he still can do what he loves. He didn't suddenly have to work with/through machines now.
There might be a lot going on, depending on what kind of programming tasks you use AI for, but here's one hunch: I think the current AI usage pattern disrupts our core sense-making mechanism on the code we create. It jumps from nothing to a full implementation too quickly and there's no gradual learning curve (impedance match) gently pulling us along, which makes the coding process highly disorienting.
how are we supposed to answer that? we don't know anything about you lol. you gotta give me more info, like, results for what? when you're coding? what are you working on? is it driven by yourself or is it for your job? help me help you.
Choose between 1. You don't love what you're working on or 2. You don't enjoy working with AI. IMO those are the two primary problems, and both solvable in their own way (more or less).
Same tbh. I lost my motivation to do any work aside from my job for a while. Now I'm working on some stuff for my friends, heavy using AI and it's just...meh. Whatever feeling of craftsmanship and skill I had is gone, now it's mostly mechanical. And if this is the case, I feel like I might as well go do something more interesting / lucrative / beneficial to society.
First, anhedonia is a real psychological crisis. You might want to speak to a professional.
Second, the old saying "it's the journey not the destination" is apt here. If you didn't enjoy the process in the first place then even if AI never came around something would have knocked the wind out of your sails. If you love the process what do you care that AI can do it faster? You can still do the thing. Either learn to love the process or begin the journey of searching for some other thing to fill the void left behind.
If you can't resist the temptation to pull the AI slot machine lever (which is not necessarily a personal failing!), you might want to seriously examine what your actual motivations are. If you have to use AI for your day job, then yeah it sucks the journey is being taken away from you.
Yes pretty much have to do it for work or at the very least deal with AI PRs and reviews. Also it now feels kinda silly to do it manually even for personal projects because I know there is a faster/better? way and before I wasn't doing it to be hardcore I was just doing it because that's what needed to be done now doing it by hand is like people who enjoy working with those old punch card computers which is a thing but not really the same. The broader ecosystem where you could get help like stackoverflow is also disappearing so I am not even sure how you would solve tricky undocumented issues now maybe by digging really deep into the systems but doesn't seem practical.
This is one thing that has really hit me: I have a question and I always default to Gemini flash or haiku instead of finding quality writing on a subject. So I get the same tinned bullet-point format every time, and my knowledge has become wider, but can only be so deep--only so many things can fit in one's head at any time--so I pay the opportunity cost of finding something that I enjoy sinking my teeth into.
I think about my favorite bands, or games, etc. the ones I cherish most are those hidden gems that I either stumbled upon during some rabbit hole or as a recommendation from a friend. How do I get that now? I can ask Claude, I guess.
If you do try to go down rabbit holes with AI, you get a deluge of information, but it is basically like what Google has become, minus the ads: aggregates the most easily accessible data, blends it up and distills it to some slurry that might kind of be coherent, and which will definitely sound authoritative, but... Is it?
And, of course, then what happens to the hidden gems that have a very low (whatever the equivalent of) PageRank but maybe because they're not highly optimized, ultra-processed and generally agreeable or popular. The channels through which one could readily receive information are all being flooded out/absorbed. Maybe I should try Kagi, or get into irc and build that rss graph.
Two's complement just had a great episode on this topic. None of us have absorbed the costs of the tradeoff we are all making by using these things, much like we didn't for using social media, or the slide into the techno-corporate surveillance state that happened after the patriot act. We are paying those costs and I am certainly feeling them more than ever recently.
It seems most companies who employ developers are going the direction of agentic coding, so for most of us there is not an option. This is where the real crisis is.
> If you love the process what do you care that AI can do it faster? You can still do the thing
I can still hike to the top of mountains even though helicopters exist, but if helicopter tourism is as popular as it seems to be becoming, and Bluetooth speakers are as available as they are, the journey and result is going to feel very different, if it's not entirely ruined.
The comparison breaks down because noise pollution is unavoidable. AI hype might be unavoidable unless you code in a vacuum but you can be disciplined in how much you let it in. Even if AI cheapens the actual output, it fundamentally cannot touch your intrinsic motivations for work, which was my original point.
Isn't that the actual endgame, what we're doing? All tasks done, all software written, everything automated, housing and food for everyone, humanity free to enjoy whatever?
So yeah, enjoy whatever makes you happy. Take a break, find a hobby, travel somewhere. Or do nothing, that also works.
It's in the act of solving greed and cruelty right now, but it's gonna magnify both for a little while before it's finally solved forever. There won't be any humans left alive when it's done, but that's the price we must pay for "Peace on Earth".
When using AI you're not the one doing the work, thus you don't feel the same satisfaction as crafting something yourself, nor does it incite any passion.
I had the same feeling at some point, until I went back to coding by hand just to exercise my skills, and I have to say after a while it felt so much better and I was hooked on working my little side project. Using AI doesn't scratch the same itch for me.
Many people who enjoyed building and coding things are going through this right now, myself included.
In my case, I felt that I was pretty good at rapid product prototyping and had made that part of my identity. Then AI agents made it trivial. It took me some time to process and accept that reality. But I’m okay with it now. I simply had to move on.
After a while, I found two things that have helped:
1. Leaning into other areas where AI can boost my skills rather than just replace them, such as writing (for example, my last two posts made it to the front page of HN) and sales (leads, proposals, etc)
2. Taking on more ambitious projects. I used to think about side projects I could complete in days, and now I think more about projects I can complete in months. With AI, days have become hours and months have become days, so I simply try to think of projects that would have taken me a long time before.
Me, programmer by trade (well, before I retired). But now I use LLMs to help me with electronic (hardware) projects. As a hobbyist (and a somewhat dim bulb when it comes to electronics) it's been a pretty good match. I've been enjoying creating projects that were perhaps just out of reach before LLMs. (Either that or they are a bit improved for having the LLM over my shoulder.)
Because I'm guessing your passion was for the craft and not just the result.
Before there were factories, if you wanted a chair you needed an artisan to make one for you. When factories pumed out chairs, there was no need for the artisan to make chairs. But building factories that made chairs was a craft in itself. If you wanted a factory to make chairs, you needed an artisan to set one up for you. A craft became redundant, but enabled another to exist.
It's natural to mourn a craft that becomes redundant. But such an event is not unique.
I do think there is something different about AI, though, in that it attacks the craft of cognition itself. At some point there's no room for a human artisan to build on top. The required faculties become too great for any human mind.
> there is something different about AI, though, in that it attacks the craft of cognition itself.
It's particularly frustrating that it's simultaneously superhuman and incredibly unreliable when it comes to cognition.
E.g. it can whip out 100,000 lines of code in the time you'd write 500...but 5% of that code will be janky, or subtly buggy, etc.
It's the "oh...you're absolutely right!" effect. I can't think of any other tools that are so powerful yet so finnicky and almost guaranteed to fail in ways that are so difficult to detect.
Try to increase the complexity of the work you're doing. It's not that AIs don't makes mistakes but they are able to pattern match to larger and larger pieces of the problem as more people use it and more iterations of the model are released. So your challenge is now in steering it in a way that these errors are minimised while reducing bloat.
Try to think of yourself as a professor who's trying to come up with a problem statement worth solving. The AI is your "lab" that will help you run experiments.
Your brain, your values. Only you can know why you ascribe less value to things that AI readily produces, but it is fundamentally a choice you get to make.
In the beforetimes, a lot of people didn't value software, but many software developers still did.
You don't have to jump through hoops of your own making to justify the things you want to enjoy and, outside of the workplace, people won't value you as an individual based on any set of performance criteria.
Programming was the thing that clicked for me most out of anything I've done in my life. That's not to say I was an exceptional programmer, probably a bit above average among professionals but not by much. However for some reason I am absolutely having the time of my life using AI tools to build all kinds of fun and useful (to me) things.
I do sometimes miss programming by hand; the flow state, the satisfaction upon completion, the sense of ownership and identity. But it feels similar to how I look back on doing math in college... a fun thing I was pursuing at the time. An academic and pure thing that I may return to as a hobby someday.
91 comments
[ 0.22 ms ] story [ 15.4 ms ] threadI wonder how carpenters felt when powertools started taking off?
I think the more apt comparison would be a carpenter to a mass produced factory. If you've read any of Yegge's recent ramblings than factories of factories seem to be what the AI focused peeps like to produce.
I'm sure there's some survivorship bias at play here. You don't hear what the tradesmen who shied away from using powertools thought about them -- because they ceased to be tradesmen.
In the same way, in the future, all software engineers will be raving about their favorite coding agent.
I dunno, I feel my comparison is more apt. AI tries to anything & everything. A powered drill just drills, and saves your arm/wrist from a manual drill.
I wonder if hand coding will become mostly a hobby and maybe a niche market. I’m not saying I like it, that’s tbd, but I wonder about it.
AI feels like something else entirely. You're not even doing the writing or the thinking, you're telling someone else to do it and they "do" it.
It's more like a carpenter walking into a factory and saying to someone I want to build xyz and it builds it immediately.
Cut that out and you're just managing an email inbox.
User expectations are also going to grow. If you're already a dev, you're already ahead if you keep up. It's always been the case that someone could release a free package that makes your personal proprietary software suddenly hard to justify.
I'm not surprised professional software developers are not enjoying AI development.
You stopped creating value and started extracting it. You stopped thinking about how to best serve humanity and started thinking about how to use AI
The model I think makes sense here is the Csikszentmihalyi Flow model, by another name the challenge-skill model of effort: https://en.wikipedia.org/wiki/Flow_(psychology)#Characterist...
Roughly speaking, you might have taken a task that was in the control, flow, or relaxation zone, closer to the center there, and lowered the difficulty so much that now it's flat out in the middle of the boredom zone.
My answer to this is to either seek out much harder problems (moving it back up into challenge, for example), or to change to a task where I have lower skill and so am in the learning mode rather than the execution mode, which I find much more pleasant and compelling than boring.
You change the goalpost, basically, or give yourself a handicap, to bootstrap yourself back up into motivating regions of the graph. Just a model, but I think it's one that is really useful.
When I was little, I was surprised to find that using a Game Genie made a lot of Nintendo games instantly become boring.
We are all Lee Sedol.
https://woventhought.substack.com/p/ai-assistants-need-adapt...
3) showing off your complex hand-crafted accomplishment no longer earns you as much social capital as it did five years ago.
Second, the old saying "it's the journey not the destination" is apt here. If you didn't enjoy the process in the first place then even if AI never came around something would have knocked the wind out of your sails. If you love the process what do you care that AI can do it faster? You can still do the thing. Either learn to love the process or begin the journey of searching for some other thing to fill the void left behind.
I think about my favorite bands, or games, etc. the ones I cherish most are those hidden gems that I either stumbled upon during some rabbit hole or as a recommendation from a friend. How do I get that now? I can ask Claude, I guess.
If you do try to go down rabbit holes with AI, you get a deluge of information, but it is basically like what Google has become, minus the ads: aggregates the most easily accessible data, blends it up and distills it to some slurry that might kind of be coherent, and which will definitely sound authoritative, but... Is it?
And, of course, then what happens to the hidden gems that have a very low (whatever the equivalent of) PageRank but maybe because they're not highly optimized, ultra-processed and generally agreeable or popular. The channels through which one could readily receive information are all being flooded out/absorbed. Maybe I should try Kagi, or get into irc and build that rss graph.
Two's complement just had a great episode on this topic. None of us have absorbed the costs of the tradeoff we are all making by using these things, much like we didn't for using social media, or the slide into the techno-corporate surveillance state that happened after the patriot act. We are paying those costs and I am certainly feeling them more than ever recently.
I can still hike to the top of mountains even though helicopters exist, but if helicopter tourism is as popular as it seems to be becoming, and Bluetooth speakers are as available as they are, the journey and result is going to feel very different, if it's not entirely ruined.
So yeah, enjoy whatever makes you happy. Take a break, find a hobby, travel somewhere. Or do nothing, that also works.
This obviously won't happen, because we could do it now, could have done it already, long before AI.
Let me know when AI solves greed and cruelty.
It's in the act of solving greed and cruelty right now, but it's gonna magnify both for a little while before it's finally solved forever. There won't be any humans left alive when it's done, but that's the price we must pay for "Peace on Earth".
Ocean view mansions and steak dinners for everyone!
I had the same feeling at some point, until I went back to coding by hand just to exercise my skills, and I have to say after a while it felt so much better and I was hooked on working my little side project. Using AI doesn't scratch the same itch for me.
In my case, I felt that I was pretty good at rapid product prototyping and had made that part of my identity. Then AI agents made it trivial. It took me some time to process and accept that reality. But I’m okay with it now. I simply had to move on.
After a while, I found two things that have helped:
1. Leaning into other areas where AI can boost my skills rather than just replace them, such as writing (for example, my last two posts made it to the front page of HN) and sales (leads, proposals, etc)
2. Taking on more ambitious projects. I used to think about side projects I could complete in days, and now I think more about projects I can complete in months. With AI, days have become hours and months have become days, so I simply try to think of projects that would have taken me a long time before.
Before there were factories, if you wanted a chair you needed an artisan to make one for you. When factories pumed out chairs, there was no need for the artisan to make chairs. But building factories that made chairs was a craft in itself. If you wanted a factory to make chairs, you needed an artisan to set one up for you. A craft became redundant, but enabled another to exist.
It's natural to mourn a craft that becomes redundant. But such an event is not unique.
I do think there is something different about AI, though, in that it attacks the craft of cognition itself. At some point there's no room for a human artisan to build on top. The required faculties become too great for any human mind.
It's particularly frustrating that it's simultaneously superhuman and incredibly unreliable when it comes to cognition.
E.g. it can whip out 100,000 lines of code in the time you'd write 500...but 5% of that code will be janky, or subtly buggy, etc.
It's the "oh...you're absolutely right!" effect. I can't think of any other tools that are so powerful yet so finnicky and almost guaranteed to fail in ways that are so difficult to detect.
Try to think of yourself as a professor who's trying to come up with a problem statement worth solving. The AI is your "lab" that will help you run experiments.
AI does not create world shattering products.
You haven’t made any big hits either (with or without AI).
What’s changed?
Use AI all you want, that isn’t going to bring your passion to life. It’s no different than using an IDE or React or GitHub.
The world is still starving for novel experiences and solutions to annoying problems and clunky workflows.
AI hasn’t changed that. It hasn’t delivered anything either. For that you need the lonely mind of a man
In the beforetimes, a lot of people didn't value software, but many software developers still did.
You don't have to jump through hoops of your own making to justify the things you want to enjoy and, outside of the workplace, people won't value you as an individual based on any set of performance criteria.
I do sometimes miss programming by hand; the flow state, the satisfaction upon completion, the sense of ownership and identity. But it feels similar to how I look back on doing math in college... a fun thing I was pursuing at the time. An academic and pure thing that I may return to as a hobby someday.