Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

The Tower Keeps Rising

lucumr.pocoo.org

Post reply on HN
Print view View original

171–180 of 280 posts

Re: The Tower Keeps Rising

#171

Post by jambalaya8 »

ColdStream wrote: post #158
Animats wrote: post #13

Earlier quoted context omitted.

The upper bound on program complexity used to be the power of the human mind. "Vibe coding" can break through that barrier. But not because the problem being solved needs that complexity. Because the process does not drive itself towards compact abstractions. It's the AI-powered version of the scaling problem Brooks described back in "The Mythical Man-Month". The combinatoric problems get worse with scale. Concretely…

The question is that if it can go beyond the upper limit of human mind complexity, would we be able to understand it or use it? I cannot remember who it was but there was an author who was traveling with their dog. They noticed that their dog would always pee on various tree to mark them as their territory. On their travels they ended up need some giant Red wood trees and figured, "I want to give my dog the ultimate…

Excellent post.

Re: The Tower Keeps Rising

#172

Post by noisy_boy »

One thing that I have noticed somewhat cuts through the haze of prompt->change->prompt is to drop into the editor when I notice something that is not exactly bad but isn't exactly what I want. It doesn't even have to be a big change - infact smaller annoyances are better in this case because they itch more.

Don't let the agent scratch that itch - it is your hungry inner programmer expressing the need to eat i.e. do it correctly as per your taste. That morsel of work is sustenance.

It also activates the thinking gears that only kick in when I am writing the code. In the same process, the code sinks in better and becomes part of the mental model, even if I may not have paid attention to every single line.

I have also noticed that the mode switch puts me in a marginally better mood - sort of a recharge for the next bout of prompt->change->prompt.

Re: The Tower Keeps Rising

#173

Post by jaza »

HiPhish wrote: post #72

> But it’s not the biblical story. At Babel, the loss of common language stops construction whereas in AI-assisted engineering, construction can continue after shared understanding has already collapsed. The lack of an immediate failure is what makes it curious and a bit disorienting. The tower does not fall, and so we do not notice what was lost. It just keeps rising. I don't know whether the author thinks this is a…

Yes!

https://thebullshitmachines.com/

Re: The Tower Keeps Rising

#174

Post by travisa »

robocat wrote: post #104
HiPhish wrote: post #72

> But it’s not the biblical story. At Babel, the loss of common language stops construction whereas in AI-assisted engineering, construction can continue after shared understanding has already collapsed. The lack of an immediate failure is what makes it curious and a bit disorienting. The tower does not fall, and so we do not notice what was lost. It just keeps rising. I don't know whether the author thinks this is a…

> Intelligence is knowing that a tomato is a fruit Intelligence is learning to avoid using childish cliches, unless your intention was to mislead. Categorisation and understanding dependencies are hard enough problems already. At the supermarket or taxation (Nix v. Hedden) tomatoes are a vegetable.

Apparently, intelligence is to deliberately misunderstand the point being made

Re: The Tower Keeps Rising

#175

Post by rtpg »

> But it’s not the biblical story. At Babel, the loss of common language stops construction whereas in AI-assisted engineering, construction can continue after shared understanding has already collapsed. The lack of an immediate failure is what makes it curious and a bit disorienting. The tower does not fall, and so we do not notice what was lost. It just keeps rising.

I don't think we have hit a long enough timescale to say that we _can_ actually build on top of the AI-assisted engineering work. Tech debt tends to reveal itself later on in the process.

Personally I've totally witnessed AI-assisted work that is clearly heading towards a brick wall quite quickly. I think "mixed feelings" is my optimistic view.

People really discount how well _well engineered abstractions_ are required for AI tooling to not generate absolute garbage. And if people are relying on AI tooling for the abstractions... I think you're really tempting fate in the mid-term.

Re: The Tower Keeps Rising

#176

Post by CamperBob2 »

irishcoffee wrote: post #137
CamperBob2 wrote: post #75

Earlier quoted context omitted.

I'll meet you in the middle: an LLM "understands" words in the same way a toddler understands the phrases they say. How'd your toddler do at IMO last year?

Ask an LLM for the difference between a definite and indefinite article, and then pronouns. Maybe just a whole run down of basic grammar. And then read the comment again.

Why don't you do that, since you're the one who has no idea what will happen?

Re: The Tower Keeps Rising

#177

Post by irishcoffee »

CamperBob2 wrote: post #176
irishcoffee wrote: post #137

Earlier quoted context omitted.

Ask an LLM for the difference between a definite and indefinite article, and then pronouns. Maybe just a whole run down of basic grammar. And then read the comment again.

Why don't you do that, since you're the one who has no idea what will happen?

I was referring to you referencing “my toddler” but I suppose if reading is not your forte suggesting you use an LLM isn’t helpful. My mistake.

Re: The Tower Keeps Rising

#178

Post by csomar »

HiPhish wrote: post #72

> But it’s not the biblical story. At Babel, the loss of common language stops construction whereas in AI-assisted engineering, construction can continue after shared understanding has already collapsed. The lack of an immediate failure is what makes it curious and a bit disorienting. The tower does not fall, and so we do not notice what was lost. It just keeps rising. I don't know whether the author thinks this is a…

You picked a bad example because this is a thing where LLMs excel. A human doesn't know that tomato doesn't play well in a fruit salad until he learns it; not that different from LLMs.

What LLMs lack, from my experience, is two things: First, a long term memory (even a zoomed out one) and Second, the ability to execute a multi-step task without fizzling out. Upon adversity, LLMs breakdown and start making shit up. The newer models are better (and that's the difference between Fable/5.6 and GLM 5.2) but still have a very limited ability to execute any tedious planning tasks.

Re: The Tower Keeps Rising

#180

Post by what »

stavros wrote: post #105
Avicebron wrote: post #101

Earlier quoted context omitted.

Do you have specifics? It would be interesting to see what kind of improvements are possible.

I just see in my usage that I can release tens of features a day, whereas I'd be able to release one or two a day usually. I don't know if it's 100x, but it's definitely more than 10x. I've written up my process here: https://www.stavros.io/posts/how-i-write-software-with-llms/ The biggest thing to get right is to let the LLMs do what they're great at (code implementation from very detailed specs, and code review), a…

Can you show us the 10s of features you’re releasing daily now?
Post reply on HN