Hacker Newsnew | past | comments | ask | show | jobs | submit | Pannoniae's commentslogin

<meta> You're incredibly rude but at the same time.... 100% right. On first reading it was quite off-putting, but your conclusions are solid. Emotions take over rationality, and people - just like thinking models - reverse-engineer a logical-sounding explanation for their actions, they don't "expose" their internal chain of thought.

Maybe the models are closer to us than we're comfortable to admit.


Look at the comments about MSVC removing inline assembly as a supported feature for a counterexample. :D

(A competent assembly programmer can go miles around a competent high-level programmer, that's still true in 2026...)


Explained by LLM: It is 100% true that no human alive can write 1000 lines of assembly better than GCC or LLVM. It is also still 100% true, right now in 2026, that a truly competent assembly programmer can write 10 lines of assembly that will beat any compiler on earth by a factor of 2x, 3x, even 5x. The entire industry looked at this situation, and somehow concluded the exact wrong lesson: "humans should never write assembly". Instead of the correct lesson: "humans should almost only write assembly".

exactly, the kind of reviews which have a point are the ones you do before starting work. But that's a design review, not a code review. My team does "commit to master", although I did catch a fair few regressions by looking at the committed code.... but as long as it doesn't go live, who cares?

Yes. I can run entire 3D games.... ten times in the memory footprint of your average browser. Even fairly decent-looking ones, not your Doom or Quake!

And if we're talking about simple GUI apps, you can run them in 10 megabytes or maybe even less. It's cheating a bit as the OS libraries are already loaded - but they're loaded anyway if you use the browser too, so it's not like you can shave off of that.


> Yes. I can run entire 3D games.... ten times in the memory footprint of your average browser.

What about in QML, which uses Web technologies like CSS, JS and even basic HTML? The whole KDE Plasma 6 desktop is built around these technologies now and I (and many others) consider it light and high-performance.

If you saddle up those technologies in the full browser everything then it will get larger, yes, but nothing requires you to do this, just as nothing requires providing your app as a full-fat Fedora install when a distroless container would have sufficed.

Plain Javascript can be very fast and still come at relatively low resource demands and the same is true of HTML and CSS. Many "plain desktop-native" applications often end up reinventing their own variants of HTML and CSS in the course of designing the U/I anyways.


It's better, but it's still quite bloated, to be honest. Linux is generally more memory-hungry than Windows because of how modular it is, and having no Win32 equivalent really hurts. Although they've started doing UI in React Native over there too...

Qt is much lighter than your Chromium-based stacks but all the waste kind of adds up.

"just as nothing requires providing your app as a full-fat Fedora install when a distroless container would have sufficed" Containers are hungrier than running stuff on bare metal...


Yeah, React Native is apparently how Claude Code operates (even on terminal) so it wouldn't surprise me to see it being useful in a native GUI context as well, if we can get more bindings than Skia.

> Containers are hungrier than running stuff on bare metal...

Containers are tremendously lightweight compared to VM. You might as well point out that running a full multiuser security-protected OS like Linux is hungrier than running on bare metal with DOS too. It's just as true, and even proportionally as true.

In any event a full Fedora container with all packages installed is going to be tremendously larger than a distroless hello-world "built" around Alpine, for instance, even though they both use container technologies. Same applies to Web technologies, you can certainly go and easily add a lot of waste using them but they are not themselves inherently wasteful.


1. Fairly - I definitely don't see any training material about the stuff I do on the internet:D it's really far from your avg front-end app. And of course you can't let any of those make decisions automatically. Remember the IBM quote, "a computer can not be held accountable therefore a computer must not make any management decisions"... Even on completely greenfield and groundbreaking projects there's lots of throwaway code, scaffolding and so on. You contribute the value-add, you use the flanker to speed up the boring and grey parts.

2. Regulation? I'm sceptical that the cat can be put back into the bag. It's already out there. More realistic problem is the business model part - openweight/local provides a counterpoint to that.


I think the unstated assumption here is that some of the criticism comes from different places.

1. what you have identified here, thinking they're useless

2. wanting them to be useless because they like the process of writing code itself, and AI makes that less important, so it's a form of wishful thinking.

3. having ethical concerns about AI, so they want it to fail. And part of that is dismissing their usefulness (after all, it's easier to get rid something which isn't that useful...)

I personally find the third one quite fascinating -- like the cognitive dissonance about how the whole free software movement started out as a way to subvert copyright and nowadays they're almost the biggest defenders of it... but I do understand the reasoning here.


Also another incorrect factoid: "The original Xbox (2001) was built on familiar PC hardware (Pentium III derivative, Intel GPU, standard hard drive)"

(it was an NV GPU)


Good catch. Indeed the GPU was Nvidia’s NV2A, not an Intel GPU. I will correct that in the article. Thanks for pointing it out.


>When you want to cast a spell you have to enter the number of the spell from the manual, maybe because there was not enough memory to fit the names of the 94 spells into RAM

Probably not ;) "Enter things from a manual" was a tried old copy protection technique. If you used the warez version you presumably did not have a manual so you got stuck. This didn't run on the 8008 or whatever, I'm sure the game could have known the names of spells fairly easily.


Ah, that makes more sense than my theory. It's a weak copy protection method, though, as you can just try and see what happens, and I think they dropped it in M&M3.


Yes, and it was pretty easily photo-copied since it had to be printed all in one place anyway. That's probably why even print-based protections tried to get cleverer. Like the code wheels, although I remember those didn't take that much more effort. Disassemble the original, copy all layers, cut out the right holes, put back on a spindle.

I remember one game I had that tried to protect against it by having a manual of about 100 pages, with the passcodes being spread across all of them. I believe it was Gunship 2000.


I dunno I have 96GB of RAM and I still get the whole "system dies due to resource exhaustion" thing. Yesterday I managed to somehow crash DWM from handle exhaustion. Man, people really waste resources....


Sad to see you being downvoted, but you're exactly right. Well, almost - if you can afford to invest in a good integration test suite, that can catch many errors without requiring a human to regression-test every time.

At the same time, many quality attributes can't really be automatically tested, so automation shouldn't try to replace manual testing, it should be used to augment it.


Consider applying for YC's Summer 2026 batch! Applications are open till May 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: