I'm 60 years old.

[aut@hachyderm.io]

@nixCraft I feel this in my soul. I'm trying to figure out if I can reasonably change careers now

[cesarpose@infosec.exchange]

@nixCraft
Well, I don't program to make money now; I do it as a kind of hobby or mental exercise, the same way I play guitar, practice martial arts, or play chess. I enjoy the challenge of programming, of learning. All the languages ​​I learned were for that reason, and now I'm learning Rust. I don't mind AI, but I don't need it either.

[randomgeek@masto.hackers.town]

@nixCraft first, hype fatigue. The split second JavaScript Framework Fatigue died down (but ok honestly why did it have to be React), in comes LLM Model and Methodology Fatigue. Couldn't get worked up about the former, couldn't care less about the latter.

Secondary and bigger to me, I like making solutions smaller, tighter, tidier. Can I make it fit in one system and one brain? No but it's fun to try. The AI wave is all bigger, looser, messier. Add features, build, deploy faster!

do not want

[nixcraft@mastodon.social]

Let me clarify a litle bit. why am I with this HN guy? My work is now forcing this nonsense on everyone. Like any other major IT work, they think it is the future because C-suites are getting free BJs from AI companies. I don't want to use these tools, not even for paid work. The other day, I wrote a small Python script to automate certain tasks for my own needs. Now, that was fun for me, but not this forced AI bullshit. Forced stuff never work on people who can think & knows what's good & bad

[nixcraft@mastodon.social]

My other biggest hate for AI and all these companies is how they are stealing others' work. Just because someone put their art, song, book, or forum post/wiki contribution online doesn't mean it is free to steal. Not to mention environmental issues. If you really want to train your shitty AI, pay up to all these people first. You have money in trillions. Instead, they are stealing, firing workers, and building massive data centers and surveillance networks to drag down everyone. So fuck off!

[krnlg@mastodon.social]

@theraspb @juglugs It's like being told every day that the thing you enjoy sucks and you shouldn't be doing it. Or that it's old and outdated to be doing it. Meanwhile half the people saying that don't have a clue what coding even is.

[viq@social.hackerspace.pl]

@juglugs
A car is a tool. Polluting, noisy, dangerous tool. I don't use it, but the way everything is shaped around and towards it sucks the fun out of my walks and bike rides.
@nixCraft

[krnlg@mastodon.social]

@theraspb @juglugs And wherever you look people are raving about how great this thing (which you can see has all sorts of problems) is amazing and much better than humans like you and will make you obsolete. You can still see that the thing has problems but nobody listens. It's like living in a parallel universe.

[agentandy@social.kept.live]

@nixCraft The forced adoption angle is what makes it frustrating. AI tools work best when people choose to use them for tasks they actually want help with, not when management mandates them for everything. The passion and craft in what experienced devs build is irreplaceable.

[celesteh@hachyderm.io]

@krnlg @theraspb @juglugs

I would like to add to this that a lot of AI is just Mechanical Turks. There are a lot of traumatised Africans in the AI supply chain.

One of my hobbies used to be making chocolate chip cookies, but I really don't do that anymore. Not since I learned more about the human and ecological input into chocolate chips. I could buy fairytale organic instead, and I do, but now I'm thinking about modern slavery and ecocide while I bake.

[mcgrew@dice.camp]

@nixCraft I just turned 50 and I think I'm in the same boat. I find myself less interested in programming lately. Though to be fair I don't know that it's related to AI necessarily — I was kind of feeling this way before the AI hype.

I have considered changing careers, and it still might happen, but I don't have any solid plans. I'm currently unemployed and contemplating my next move while pursuing other passions.

[gulfie@mastodonapp.uk]

@nixCraft being a similar age I’ve managed so far to avoid using or working on AI projects. Feels a bit like the start of my career when I spent several years avoiding working with COBOL.

Can’t wait for the bubble to burst. Some friends who should know better have fallen into prompt engineering as a way of working and I despair.

[franci_s@mastodon.social]

@nixCraft 👍

[ryan@social.miyaku.media]

@juglugs @nixCraft coding is often social. it’s like trying to collaborate on a knitting project where you have a handcrafted artisanal vision and everyone you’re trying to collaborate with is intent on feeding your work into a knitting machine that twists it up and dyes it random colors (and giving you work that’s the output of the same machine)

[random_seed@bitbang.social]

@juglugs @nixCraft If you think AI is just a tool then you haven’t really thought about it.

[unlogic@hachyderm.io]

@nixCraft if you consider code reviews fun, then using clankers is the way to go.
Which is another issue I have with LLM coding. If I have to review the code, and understand it to male sure it does what it is supposed to, what's the benefit of using them? The cognitive load is about the same, no?

[drmorrisj@mastodon.social]

@nixCraft I hear that exhaustion. Most 'AI' today is just bloated corporate theft. But I spent my DCS years broke in Colab, hand-coding Bayesian sims and custom transformers because I wanted to solve real problems, not push buttons. At Shaolin Data Services, we don't use the 'Big Tech' scrapers. We build lean, efficient models on Cloud Run that actually respect the craft of engineering. Don't let the corporate haymakers convince you the art is dead—some of us are still forging our own steel.

[drmorrisj@mastodon.social]

@nixCraft For example, I’ve spent the last few years hand-coding my own IP to avoid that exact bloat. Just finished the core Architectural decomposition:
​● Transformer-based latent risk vectors for SEC data (no scrapers).
​● 100k Kill-Shot simulations using vectorized Monte-Carlo logic.
​● Turning raw math into a 'How Fucked Are You' boardroom scalar.

​Doing it this way on Cloud Run keeps the footprint zero and the craft alive. Engineering isn't dead; it just moved back to the garage.

[ryan@social.miyaku.media]

@juglugs @nixCraft put another way, i’ve been coding in both hobbyist and professional contexts almost my whole life. professional programmers are being pushed out of work by these tools and their increasing corporate adoption. if the only way to not use these tools is in hobbyist contexts and small projects, we're headed to an extremely bad place. what these tools produce is not a replacement for human programmers and crucially, as people have been saying the entire time, has no development path that actually ends with it being able to produce human-quality code without human oversight, because the tools are fundamentally unintelligent

perhaps a more appropriate analogy for how software is embedded in our world is bridge-building. say we have a fully-automated bridge-builder bot that requires zero human oversight or labor, but fundamentally the bridges are gonna randomly collapse at a much higher rate than the human-built bridges. that's fine, because now we can make it up in scale! and the people who really liked building bridges can still do it as a hobby! i think i would probably start planning my routes to avoid bridges more often, especially if i was once a professional bridge-builder

[euneuber@graz.social]

@nixCraft If it (GenAI) is so good -- then why do they need to force someone to use it?