gloosx

can't wait to beat it with a face-swap or some random driving license found on the internet

I like how it accumulated 3 such replacements before finally getting fixed as a reaction to this comment with PR 14543[0], but after review, two "fix unit tests" commits were added, one of which replaces claude with copilot and the second one messing up the docs markdown, getting merged after that. Agentic workflows are the battle...

https://github.com/github/gh-aw/pull/14543

A novice prefers declarative control, an expert prefers procedural control

Beginner programmers want: "make this feature"

Experienced devs want: control over memory, data flow, timing, failure modes

That is why abstractions feel magical at first and suffocating later which sparks this whole debate.

Natural language is not the highest level – it is just the highest level that still needs words.

earth-moving machines don't really move earth because they are told so, machine requires fine-grained deterministic input in a certain language of motion to do what it does.

imagine if machine operator was saying "move this earth here, not not this one, 30 cantimeters to the left... yes, move this one earth here, no not here, please a bit to the right", lol

of course, someone with no exprerience with the machine would prefer telling it what to do, and he will even achieve something rough, but someone who knows what handles and pedals do would prefer the handles and pedals

It's not like we are 100% sure, it's possible a real human would be writing like this. This particular style of writing wasn't as prevalent before, it was something more niche and distinct. Now all the articles aren't just looking like a fairly sanitized blog posts - they are all looking the same.

Wow that was my first impression as well. Is this the new norm for articles to be all same?

All these bullet points; This was not X. This was Y Verdict was not X. It was Y. Markdown isn't X. Markdown is Y. Malware doesn't X. It does Y. This wasn't X. It was Y. The answer is not X. The answer is Y. If an agent can't X, it can Y. Malicious skill isn't X. It's Y. Full stop.

I would rather read the prompt honestly

But there is a non-zero chance it will not stand straight 100% of the time? Pets, earthquakes, human shoulders, you name it, something can throw this thing off balance. Plus, there are people higher than 6'3", people on the ladders, children on the shoulders, and they all can pretty much look at it directly from above.

Millions in marketing efforts? Anyways, it may be a key part in generating code, but that was always a lesser part of software engineering. If it's generating code it doesn't mean it is doing any engineering for you or becoming a "key part" of it in any way.

I would be quite scared to have such a monster lamp near me indoors. I mean, 60k lumens can cause temporary blindness, disorientation, and potentially retinal damage. It's near the welding arc territory at this point.

>Not once has it ever changed on me.

I don't know how did you achieve it, but I was doing it countless times.

Open with -> other -> enable all applications -> always open with.

For a short while it works, but somehow, something always reverts it back to xcode. Maybe it is restart. Maybe it is little evil cron job discreetly changes it back to xcode, but I was never able to get rid of it. It is happening to me on many different machines since Sierra. One calm day I casually double-click an STL or JSON and it prompts me to install some xcode packages, and I get angry at the machine.

...or you have to be deeply entrenched in another kind of bubble to believe the opposite xD

>This requires calling native APIs (e.g., Win32), which is not feasible from Electron.

Who told you that? You can write entire C libraries and call them from Electron just fine. Browser is a native application after all. All this "native applications" debate boils down to the UI implementation strategy. Maintaining three separate UI stacks (WinUI, SwiftUI, GTK/Qt) is dramatically more expensive and slower to iterate on than a single web-based UI with shared logic

We already have three major OSes, all doing things differently. The browsers, on the other hand, use the same language, same rendering model, same layout system, and same accessibility layer everywhere, which is a massive abstraction win.

You don't casually give up massive abstraction wins just to say "it's native". If "just build it natively" were actually easier, faster, or cheaper at scale, everyone would do just that.

So, it's generative pre-trained transformers again?

So what is he even coding there all the time?

Does anybody have any info on what he is actually working on besides all the vibe-coding tweets?

There seems to be zero output from they guy for the past 2 years (except tweets)

The real bottleneck is doing it on the CPU, if you switch to WebGPU you can simulate millions of particles simultaneously in just few milliseconds per frame.

Oh boy, in 2015 Windows 10 was released, and it was extremely broken, including endless reboot loops, vanishing start menu and icons, system freezes, app crashes, file explorer crashes, broken hardware encryption and many broken drivers – so really it was about the same as now. Embracing LLMs and vibe-coding all around made this even worse of course

> The cornerstone of Microsoft still is Windows and Office.

You mean Windows and Microsoft 365 Copilot App?

sounds cool, want to explore this a bit, however, why XML if condition?

> <if score >= 10>

this looks kinda cursed, makes me wonder how it's even parsed

It's even worse when LLM eats documentation for multiple versions of the same library and starts hallucimixing methods from all versions at the same time. Certainly unusable for some libraries which had a big API transition between versions recently.

This "constitution" is pretty messed up.

> Claude is central to our commercial success, which is central to our mission.

But can an organisation remain a gatekeeper of safety, moral steward of humanity’s future and the decider of what risks are acceptable while depending on acceleration for survival?

It seems the market is ultimately deciding what risks are acceptable for humanity here

>rewriting code

Key thing here. The code was already written, so rewriting it isn't exactly adding a lot of quantifiable value. If millions weren't spent in the first place, there would be no code to rewrite.

I don't even know who the hell uses them systems for components.

Just trowing a flex-box and a few good ol' css rules does 99.999% of the job usually.

$300 for UI blocks? For what? A div with flex, gap, and padding?

Would be funny if they renamed Windows to Copilots

This is a ridiculously small sample to tell me that "the invasion had broad popular support at the beginning". It had a broad support in your own circles and you casually extrapolated it to the whole population.

According to my own relatives, friends, and acquaintances in Russia (where I'm from) – no one supports or ever supported in the beginning the total idiocracy which is happening.

>the invasion had broad popular support at the beginning.

According to whom?

You should understand that public opinion surveys in authoritarian countries are problematic. In autocracies, people might want to hide their opinions and give socially desirable answers that conform to the official government position for fear of facing repression or deviating from the consensus view.

Lithuania fought two distinct wars with Soviet Union, both conflicts involved USSR attempts to control the Baltic region and establish communist rule.

Whatever coutry has the most firepower you mean.

Hungary, Chechoslovakia, Afghanistan, Angola, Ethiopia, Azerbaijan, Lithuania, Moldova, Georgia, Tajikistan, Ichkeria, Ukraine, Syria... The list goes on

>humans are to a larger degree trained on spacial data

you must be completely LLMheaded to say something like that, lol

humans are not trained on spacial data, they are living in the world. humans are very much diffent from silicone chips, and human learning is on another magnitude of complexity compared to a large language model training

>Create a test for intelligence that we can pass better than AI

Easy? The best LLMs score 40% on Butter-Bench [1], while the mean human score is 95%. LLMs struggled the most with multi-step spatial planning and social understanding.

[1] https://arxiv.org/pdf/2510.21860v1