It's a given now that LLMs are leveraging code to do things.

"On it's own" generally means "not steered" or otherwise given professional guidance or input.

I would say "developed the necessary software for a simulator" to be even more impressive - "here solve this problem" and "OK, but first I have to built the entire lab!"

Indeed. It didn't write it's own TCP implementation either. Nor operating system, nor compiler, nor build its own wafer etcher...

My comment probably sounds facetious but I'm actually intrigued by why we seem to view using the level immediately underneath as cheating whereas the levels below that are taken for granted?

Why do we view using the level of abstraction immediately below as cheating while the levels below that are taken for granted?

Sounds like the paradox of tolerance. Being tolerant of intolerant people just gives them a platform to spread their intolerance.

Rocketjump! The wife password is four words all uppercase.

https://www.youtube.com/watch?v=bLE7zsJk4AI

I think it's not just token limits - I think it's because it's so _dense_.

You get a week of research and debugging and testing compressed into a few pages. Even if it's explained well, it's just so much information. And since it's AI, I'm constantly second guessing "is that really true?" and it's exhausting.

Weird there's no information about that here.

But if they went down in cpu usage and memory usage, removed a memory leak, and are MORE compliant with nodejs... it sounds like a MASSIVE success?

OpenRouter offers 1% discount to save your conversations, explicitly opt-in.

Anything else they don't save it. Even if they tell you the model provider saves your data for training.

Only do one new thing at a time.

Don't drive a new car, using new a new type of engine, with a new type of steering wheel, with a new type of entertainment system, .... all at once.

WHEN something goes wrong it will be intense trying to figure it out.

> My GPT-7 agent... I become .... annoyed at the AI-speak in its first paragraph

When we have GPT-7 it will still do AI-speak?!

>Asking "what have note-taking apps accomplished?" is like asking "what have spreadsheets accomplished?" or "what have cameras accomplished?". A tool doesn't accomplish anything on its own. A tool is a means to an end.

Yes, but, the _shape_ of something does offer affordances of what it makes easier or harder to do. So "obsidian" vs "saving files on my computer" does seem to be a legitimate question.

I agree with you. For example, [[links]] are a useful affordance for thinking that wasn't as popular before Obsidian. However the original piece was asking:

> Has there been a meaningful increase of understanding and creation thanks to personal knowledge management systems?

My question is: in a world where photography didn't exist, how would your life have been different? How would everyone's lives be different? It's a counterfactual that's impossible to answer meaningfully. Similarly, we can't measure how the world is different as a result of the popularization of [[links]].

From that perspective Org mode offers everything every other note taking app does.

Except a nice interface. Except syncing between devices. Except cloud backups. Except third party data breaches. Except paid support contracts.

The blog claims rust chosen mainly to address memory issues, which rust is a better language for. So, success would simply be less new memory errors / easier to patch old ones.

Claude installer/updater had a high memory usage problem (Gigabytes of resident memory used during a fresh installation of Clade with it's native install.sh script, or during a claude update). This was reported as early as December 2025, see https://github.com/anthropics/claude-code/issues/12327#issue....

The latest Claude installer (2.1.215) possibly does not have it anymore https://github.com/anthropics/claude-code/issues/4953#issuec..., but more tries on various machines are needed to confirm that.