When honesty goes out the window...

Maybe I'm wrong and I will not miss it. I feel like something is being lost though. A bit of grief is appropriate.

I love code, you don't, fine. So be it.

But as is: it just output stuff that will not scale properly, and wrangling it takes just as much time as me coding it up myself. So it's in my hands until then. I've found LLM's somewhat useful as a first-order discovery in documentation, and not much farther. Try to trust it as an oracle replacement of all search/documentation and you'll be burned once every few days.

I think you’re letting your pride get the best of you. Writing software should not be about ego. It’s about making something helpful, or something beautiful, or at the very least financially useful. Interacting with software through a natural language interface does not mandate you are creating junk. I won’t tell you how to use the tool. But I am sure if you were sufficiently motivated you’d learn how to make it work for you.

> I think you’re letting your pride get the best of you. Writing software should not be about ego. It’s about making something helpful, or something beautiful

The things I'm the proudest of are, by far, the most useless and unhelpful ones. The ones I did for myself. The ones that made me think "wow, now I rock!" and the machine spit its guts crying "mummy help me". That's what the demoscene (to name one subculture) was all about. That's what hacking is all about in the first place.

> or at the very least financially useful

These have always been the ones I'm the most ashamed about. These were the things I would not have done for myself, but for other people. The ones that kept me living, not the ones that made me feel alive. What did you think I was talking about?

> But I am sure if you were sufficiently motivated you’d learn how to make it work for you.

I'm motivated to code. I'm not an architect, a QA tester, or whatever, I'm a coder. Now, wanna know what kind of person I am and how I made LLMs work for me long before you suggest it here? Read that:

https://news.ycombinator.com/item?id=48072319#48073350

Then, come back and we can talk. I WILL NOT ASK A MACHINE SOMETHING I LOVE TO DO MYSELF, WHICH I TAKE PLEASURE FROM. If that looks like an idiotic and stubborn stance to you, FINE.

Also, humans didn't evolve to write or read code, even the best programmers are terrible at it from an objective point of view, if we compare it to what is theoretically possible.

And, no, it is not a question about style or preferences, the code is messy and contains to me as a maintainer obvious bugs.

Basically, they still need a human guiding them pretty closely unless you’re happy to just ship a mountain of vibe coded mush.

That said, I haven’t tried Astra in earnest yet. We’ll see in a few weeks whether or not it is a step change.

Just as they produce writing that isn't structured well for humans, they do the same with code. Sure, if you're willing for that code to be write-only and for agents to be the sole interface through which you interact with it, that works fine, but it is not better than what a professional, skilled human could produce. It's often full of pointless crap and design decisions that over index heavily on local structure that make very little sense when global system structure is considered.

I think the question is less about the capabilities of LLMs and their outputs (because obviously that can change), but because they represent a workflow which codifies exactly that role I mention. You fulfill the role of a manager more than a programmer. You have a vague idea of the codebase, but lack any kind of intuitive mechanical understanding. There are many cases in which I can see an issue, identify the problem, and have it fixed faster than you could prompt an LLM to investigate. After a certain point of familiarity, all reasoning is local reasoning. That's the power of mechanical sympathy.

Many, many large shops don't give a shit. They've long since galloped into the dark woods of bureaucratic know-nothing hell. Let the LLM figure that shit out. Those shops are already optimizing for maximal output and churn, so the value proposition seems obvious. I've been smugly saying it for years, but LLMs have me saying it quite a lot more these days: "Sure looks like it sucks down there!"

These companies have many billion dollar investments and yet their main product is a bloated meh.

That's a weird way of looking at it. Programming lives at the end of a well-worn path uncountably many millenia long, defined entirely by our biology. We didn't "evolve to do it" in the same way a duck didn't evolve to swim in a specific river.

Current models are in fact much better than the top models from a year ago. I have no idea how anyone could claim otherwise without a huge amount of motivated reasoning.

That's exactly how democratic governments make their decisions… you might think it's stupid, and you'd be right, but that's democracy. It's the majority that counts, not what's right. At least you can have a little fun with their arguments, they're pretty inventive you know.

Well now, he just makes me smile, not laugh. I keep my laughs to those who embrace AI yelling "hooray" that they no longer need to code while they pretended to love it for so many years. No, you didn't. You though you did, as many people think they love their partner. Or childs. But to which point? What would you sacrifice for it? Whatever you say, you don't really know, and you probably have to say it anyway for the sake of looking weird, or a so-called bad person.

I don't care of what people think here, so let's do it: I sacrified my social life to my passions. My professional life too. I turned down promotions, even early. Because not coding, or coding less, was not worth any salary. Besides I'm not made to manage teams anyway, they would blame me for being harsh, too demanding, so no, forget it. I want to remain happy, your employees too. Let me do what I love and everything will be fine (though don't take me for a grunt, I have things to say in my field, this is MY field). Yes, I sacrified my life to it. Did you? No, you're not dedicated enough. That's not a shame, maybe I'm the one to blame, maybe I'm the one pointless, the one too much this or too much that, but I am what I am.

So I won't blame him for this article. We're probably the same kind of nerds in that regard, and nerds are just that: living in another dimension. Not only different from a so-called conformity, but something more unfathomable. That's why they marginaly work together: they can't even understand each other completely.

However, I would not have written that I don't use AI. Because I do use AI, but undoubtly and definitey not the way most people do (or pretend to). And probably in a way the author did not really try. No need for the damn Claude and such, come on. Free options are enough for that way of using it. Need to refactor? Why would I ask the AI, I prefer to do it with LSP in my Emacs editor. Takes longer? Maybe. But I'm still aware of the whole thing. My brain cells refresh, like a RAM.

AI does not write my code. It often suggests, so often that it's not rare I ask it firmly to stop writting code, only talking about it, about some logic in a specific area. That's a quite different approach. And even if its code is good, I would be ashamed to kill/yank (you though I would copy/paste? Come on!). First, it's not my style, not my naming conventions, etc. I know we can lead it to use our style (users of Claude always talks about config files for such things), but I fucking don't care. I don't want to depend on this, needless to say what I think about paying for it.

I can say it now: AI is the better companion of the lonely nerd EVER. I wish the author would find it at some point. Not to write code for him, but to help when in doubt on something. Oh damn, I always have doubts in many ways. That's sane to doubt. Never leaving the thinking apart, no way! My brain cells need that.

Also to have clues of the options. Clues of the newer paradigms. Or simply chit-chat about... code. Common practices. Algorithms (more often, it just responds about things I already know, so what? It's not in my mind, let's continue). For example, it's very good to embrace modern C++, there are so many things that changed in that field. It's also good to make sense of sometimes over-verbose compiler errors, especially when you go crazy with your own templates (omg, yes, in that damn context it's a typename, such things). Or to work with unreadable regexps too. Such things again.

That's not an approach for work, for jobs, to make a life. Where we have dead-lines and must respect it as much as we can. I don't work since a year or two, I'm just keeping an eye on technology, as I always did anyway. I always work much more with my own projects, it's too satisfying to stop, or find an excuse to stop; no boredom. The best times are when I do things for myself, for my own dedication. In that context, without dead-line, the use of AI is sooo different from what they're all mumbling about. TBH, I've read many things about it, and NOT A SINGLE time I've read something that closely matches how I use it.

I don't buy time with AI. Most people do, but I definitely don't. On the contrary, I lose time... but for the better. The same way I always lost time digressing of my current goal. What's that new thing it talks about? Wait a minute... wtf? Let's dig it... wow, that's cool! One hour lost, still... not really lost. And again, and free, no charge, not for that. Yes, I need to recall some context sometimes, just enough for what a fresh session needs to know. It does not need to know all my codebase, just that bit, and maybe that bit too, and off we go.

So yes, you see, I have much more fun reading them all than reading that article which, in many ways, makes sense (and that's a breeze in the AI hype), but which is still too stubborn, and believe me, I rarely say that from other people in my loved field ;)

AI helps me learn. Not to code, I do this since ~45 years! No, learn new things, new paradigm, or old ones I may have missed, because coding field is so large you can't know everything. You just have to insist when it seems to just say what you already know. There's always something to learn at some point. One doesn't have to trust it all the way, we have tabs! Dig the docs, the manuals, the APIs... just like before, except the AI often prevents searching for too long for the good terms.

I could write more on it, a lot more, but I'm not writting an article, so let's stop now.