Aperocky

I read it twice and the second time it stood out as not.

Usually there is a taste about AI generated messages that are just quite obvious, it's usually pretty close to corporate speak, but not exactly the same.

Well, if it is being done, per the other article on the front page, we should be able to find it.

They are 20x productive on the old KPI.

And the product now iterate 2 times faster with 10 times more bugs and customers are pissed.

Looking at you Apple: https://aperocky.com/blog/post.html?slug=apple-watch-sucks

I'm just really surprised that 21% of Lynn's are boys.

claude-code is using a suspicious amount of memory and doesn't seem to care about optimizing it nearly as much as the others.

harness is not a moat. CC is great, I mix it with codex with my own vibe coded messaging and hierarchy layer. Lately there are more CC because of Opus5.5 but Codex caught back up with Sol-6.1;

Codex also wins by consuming a lot less memory.

Serious answer: A lot of times these legacy software are different from sqlite. To end to end test sqlite is a lot easier than a VPN client under distinct environments. It's simply a lot more costly and harder to setup or even possible to find all of these moving variables.

A change in behavior maybe construed as a bug, a breaking change or a regression, and a lot of time it's not so obvious. Sometimes you want the original behavior, sometimes you don't. Sometimes changing one for the better means another behavior needs to go, and you don't know if the user already depended on that other behavior. That's the real staying power of legacy software despite universal agreement that they suck.

Unless they also have 750K tests like sqlite do, I can't imagine this will go well, even if/especially if the legacy code warrants a refactor.

The problem with rewrites are that even if the rewrites are ultimately good it will tend to first create bugs on an existing thing that are presently churning in a stable state.

The better way then comes with the drawback where one vendor has all the power.

Then someone would ask "there has to be a better way" again, when the reason for the new inconvenience gets forgotten, and it oscillates.

I glossed over it, it probably deserve half the article if I was writing it.

Why would it be sad? They get to socialize with their peers, which is probably the more valuable part of the education process. Learning as I come to understand, are better curiosity driven, which don't necessarily need to happen at school. All the game hacking and number crunching on unit stats probably did more to where I am today than any school class.

Socializing (with peer) and allow parents to do that too (work is basically a form of socializing with peer, with benefits) will probably remain far far into the future, as long as humanity exist, whether AGI pans out or not.

> My wife and I are expecting our first child.

> I once tried to come up with a single-word purpose for education, and I landed upon the word “resourceful” ...

As someone who's a little further ahead in this process, I realized the true raison d'etre for education might actually not be any of the reason discussed..

It's so you, the children's parents, can go about your business for the day and the kids can stay somewhere safe and where they can interact with their peers. If they happened to learn something, that's pure bonus.

Maybe OP is on naughty list facing serious scrutiny during the postmortem review.

tmux is also a great primitive to build on top of.

You can have a message routing layer entirely on top of tmux with send keys - this way you don't even ever need to see your terminals (but can still see it if you want to).

I see your point is basically snap bad.

And as a software maintainer who publish to all distributions, I wholeheartedly agree.

The CLI takes 1GB of memory when like you said, it just "fires off to the model".

It's easily noticeable. Codex don't take half as much most times.

It looks like they are recursively gradient descending into their own local minima at increasing code velocity.

There will not be, they will find that the videos submitted are all of noise quality after the training starts, and then they would need videos of certain, specific actions that were basically not recorded. and it will go on and on for quite a long time.

This "value" you talk about, is it past tense? A single iphone probably have more compute then the entire world had in the 1970s, would have cost at that point probably hundreds of millions. You will not sell the iphone for that amount today.

There's a hidden problem you skipped over. The model may get good at anticipating your needs, but are you good enough at anticipating your needs?

Is GPT Luna 6 dethroning Deepseek V4.1 Flash? It's price seem to be undercutting flash at a relatively similar capability.

vim/past vim terminal dweller here.

I've eventually settled on a CLI agent multiplexer that essentially run in the background while the frontend is a GUI gateway agent to that backend system with a goal middle layer so I no longer need to interject directly into the prompts and forces the CC and Codexes to communicate to me in a structured format relevant to my purpose.

What makes you think that labs don't operate on "vibes"?

If there's anything that I can conclude about Anthropics idea of how a LLM should speak. Vibes would have been an euphemism

> ability to feel superior.

I actually have a random thought that it is this innate human drive that causes society to be mostly employed despite massive differences in technology and productivity across geography and time.

What if the driver of employment is not efficient return of capital, but satisfaction of human mind?

Yes, we've seen that already without necessarily a technological singularity. You accurately describe the great depression in the 1920s and several similar events prior to that.

But what's consistent with those events is that they end and economic development is resumed under different circumstances. I don't see any reason that the current era will be different or that current problem will be unfixable. People used to get paid to build national parks. What would be our current iteration of national parks?

That is one possibility but it goes against the interest of all income groups of the economy, including the ones on the very top. So a solution is in everyone's interest to be found, and nor is it a physical impossibility.

I tried to read this but it's just claude after a certain point.

It's unreadable.

I can't understand any serious writer essentially expecting other people to read stuff that they didn't write. Or if claude was actually good at formulating ideas and conducting information, fine. But the truth is that I'm just reading 80% empty prose that are forced in there by claude.