after reading so many people argue about this over the last few years (and having had my own experience - I've been writing software professionally for close to 15 years), I've come to believe people are talking past each other because different people enjoy or excel at different aspects of coding.
at the very least, there's people who enjoy the experience of hand-crafting software - typing, being "in the zone", thinking slowly through the details.
then there are others, like me, who enjoy thinking abstractly about the pieces and how they fit together. might as well be doing algebraic topology. nothing bores me more than having to type precise but arbitrary syntax for 5 hrs (assuming you've decided to use the brain capacity to memorize it), and having to fight compiler/small logic errors throughout. I like the thinking, not the doing.
yes, we havent needed AI to build this for decades. we did however need to waste a hell of a lot of time doing essentially physical, mechanical work with your fingers.
took me a week to build a poker bot in 2006 that earned me $1000 a week during college, and that was self imposed because I didn't want to raise any flags.
It was based on a world of warcraft bot that I modified, and I learned a lot during the process.
What you call wasted mechanical work I call the foundations of a career that changed my life.
Take away the AI and this guy has nothing but an idea. An old idea that has already been done to death, and none of the skills required to actually implement it and maintain it.
You might not like writing code, but that is the job no matter how many natural language layers you put on top of it.
Just to make a small addition to my comment, which also addresses the sibling and child replies.
I 100% agree that any good professional still needs (with or without AI) the "design, engineer, and knowledge of your constraints". I'm not arguing against that. Those are, in fact, part of what I find most fun about programming, and the reason why I fought through the typing boredom since I was 13. I'm also not a vibe coder.
I'm just saying all of that is somewhat orthogonal to the typing of code itself. With strong typing (as in type theory - I still write the types, sometimes signatures for interfaces, etc) and other tooling, you really can get a lot done by delegating the bulk of the implementation to these tools.
If someone makes a 3D printer for houses there is probably someone who will say laying bricks "is the real job"
It's just someone writing about his vibe coding experience. Not interesting for me, but then again I stopped reading half way and am not telling people to stop.
The 3D printer analogy doesn't hold up though, that implies design, engineer, and knowledge of your constraints.
It's more like a pachinko machine that rewards the user with house like objects that may or may not work.
If the user builds their house with it and it collapses and kills their family fine, but you can't use a system like that to build anything where you might have external liability because fundamentally you don't understand the problem domain and an ai model cannot hold a civil engineering license and be held liable for structural collapse.
so instead of learning a useful transferrable skill they talked to their magic mirror and got an output that approaches correctness without the necessary knowledge to verify it or maintain it going forward without their magic mirror.
To quote the author:
"The insane part is I didn't write a single line of this code.
All of this was created through conversations with the Cursor AI agent.
I don't even know how we got here with AI."
This is about as interesting as your average TODO MVC tutorial.
Your best example of a "historically efficient way to solve political problems" is a 4 year civil war that killed more than half a million people and, after all that, still left African Americans as second-class citizens for a century after?
The amount of violence to keep the slavery running was huge. You cant pretend that all that violence does not count. That being said, war was more about south wanting war/leave the union, because the north did not wanted to expand the slavery to new territories. That threated the south.
It is not like north would march in there to stop the slavery. There was an anti slavery army - John Brown with his, like, 20 or so people attacking south.
African Americans as second class citizens were in fact much better off then them being slaves.
We can theorize about the non-violent path to emancipation, and the speedy path to legal equality.
But it's counterfactual. It took severe violence plus 100 years to get there. Plus another 60 (yikes) to get to where we are today.
That's horrible! But nothing about that reality suggests to me that there was a less-violent or speedier way to get there. Governments are made of people.
Getting there was a worthwhile goal. I don't think there's a "but at what cost?" debate here.
So it sure doesn't feel "efficient", but it might be the "most efficient possible" in the human world.
If you want to go there, all governments and their laws (and thus politics) are predicated on their monopoly on violence, and civil society and the rule of law cannot exist without violence. Therefore all politics is violence and all political problems are also violence problems.
I do not want to go there. I made comment about huge amount of violence slavery in Americas required daily back then. Slavery was violence in amounts completely incomparable to what you are trying to equate with it.
Moreover, that sentiment was literally expressed by slavery opposition back then. Afaik, the sophistry about "any government is violence therefore, it is the same, que" was not all that much thing back then.
Slavery was just as much a matter of politics as it was violence. Separating the two as if to imply that violence can't or doesn't solve political problems is a specious argument. American politics has normalized a degree of violence in the last few months that would have been unthinkable, and the degree of violence doesn't change the nature of what politics is, only what it permits.
I've found codex to be better here than Claude. It has stopped many times and said hey you might be wrong. Of course this changes with a larger context.
Claude is just chirping away "You're absolutely right" and making me to turn on caps lock when I talk to it and it's not even noon yet.
My "favorite" is when it makes a mistake and then tries gaslight you into thinking it was your mistake and then confidently presents another incorrect solution.
All while having the tone of an over caffeinated intern who has only ever read medium articles.
The other day I was helping somebody choose some funds to invest in. The majority of "global" or "index" funds we looked into either directly or indirectly had the largest share of holdings in Nvidia, Microsoft, Meta, Alphabet, AMD. Often these were about 50-70% of their portfolio. The returns for the past year look great but the crash is going to be devastating and widespread because so many investments link back to this bubble.
There are plenty of non-US index funds, like EU, UK, Japan and others. There are also indices that track smaller companies rather than just the S&P500 or Nasdaq.
Or diversify in both directions - small US, and big and small international funds.
There are other companies than the magnificent 7. Other boring companies with boring dividends and solid earnings that have nothing to do with tech. Ulta. Pepsi. Chevron. But you sound pretty smart so I’m sure you had a good reason to invest in non us markets.
Honestly it sounds like you are missing the point while simultaneously using a bad password manager.