HN user
I click. I see an anime girlie bot challenge screen. I hear zoomer pop in ~32k 16-bit stereo. I see JavaScript animations and a seizure-inducing vibecoded layout. I close the tab.
"Miss the old web?" The author wasn't even born then.
It really does boil down to 'just read the source code bro' but with more of a sense of direction to it.
Don't worry about understanding the entire codebase just yet. Instead, try to understand some specific parts of it to begin with. As you work with different bits of the code you'll gradually form a mental 'map' of how the whole thing fits together.
Start by trying making small changes, e.g. make the guns faster, or make the player jump higher. Look for where that code is that does that thing.
You might be taken on a bit of a journey, finding the definition of a function, then the function calls within that function and the variables passed to them, etc. - all the way along you'll be discovering how various other things are done within the game engine, and the general layout/structure of the codebase overall, as well as plenty of general programming concepts that will be useful anywhere.
Do this enough times and you'll gradually build a familiarity with the engine, and programming in general. Then if there's something you want to do that's in a part of the codebase you're not familiar with, you'll be more able to 'dive in' and find where it is in the code, and you'll be more confident that you can read code you're not familiar with it and be able to figure out what it does.
I can only speak from my experience from when I was just starting out learning to code (17 years ago!), adding and tweaking various features in Cube Engine and its sequel Sauerbraten, eventually moving onto Stepmania. Stepmania was a MUCH bigger codebase than Cube/Sauer (probably still not as big as Quake 3 though!) and the stuff I picked up modding those two was invaluable for working with Stepmania.
string.h
Then why are people starving?
This is vibecoded. How are we supposed to believe that any of the descriptions aren't hallucinated?
https://shitfixer.app/s/iii7qfkywd
This is just Google Gemini with a vibecoded front end, isn't it.
The 'DEAN' and his goons in this story is a classic example of a psychopath.
The moment I entered the room, the Dean just started shouting at me.
while we were leaving, the Security Officer saw my phone was recording. and this is where all hell broke loose.
one of the security guys just held me. and started being rough with me. like pushing me around and shit.
Everyone was shouting at me, i felt like I was being harrassed at this point, and honestly, it's a little embarissing, but i just got overwhelmed. like with all this shit. I started crying, literally balling my eyes out.
Then the dean called me to his office. To his credit, like i guess seeing me cry and be so sad, the dean became really really nice.
to the dean's credit, like half an hour later, he literally came to my room, to check up on me and that i was good. like that's really nice of him, he didnt have to do that.
Textbook case of abuse. Scream at a child who's smaller than you, don't let him get a word in edgeways, physically assault him when he attempts to hold you accountable, then when he finally crumbles, act all sweet and sugary like you've done him a favour.
The dean is the real asshole in this story. It's psychopathic abusers like that who need to be removed from educational institutions. Places where they can wield their power like a weapon, abusing it against people who they know can't fight back and give them what they really deserve.
It is literally forced upon you. If you are a fully grown adult, you have every right to use a device you own as you wish. But Apple and the ever-increasingly Orwellian UK government disagree. Your device is crippled until you consent to surveillance. You are forced to hand over your identity to be allowed to use the device you bought and paid for as you wish, and now your every move on that device is linked directly to you.
The AI model is likely trained on Stack Overflow posts, or otherwise related content from wherever off the internet, so most likely it will have ripped off enough hand-written posts or articles on how to quit Vim. So of course it can regurgitate the required keystrokes on command.
And will the writers of those posts, who contributed their knowledge to help out their fellow humans, get a fraction of a penny for the hundreds and thousands and millions of dollars of profit the AI corps are making off the backs of their labour? I somehow doubt it.
Where did they get the brain cells from?
Why not, you know, bite an apple?
Looks like it may already have been patched, it's not working for me.
Seems I'm not the only one either: https://github.com/xyzeva/k-id-age-verifier/issues/7
It absolutely will not make formal verification go mainstream.
What it will make go mainstream, and in fact has already started to, is “ChatGPT verified it so it must be OK.”
I wouldn't be surprised if a fair majority of them have been taught to see goto as nothing but a vestige of the 70s which should never be used under any circumstances except as a meme or to deliberately obfuscate code.
I have recently become quite fond of goto-based error handling and find it a lot cleaner and more readable than the if-else-mountains you otherwise end up with. I just make sure to leave a comment with a link to xkcd.com/292 so anyone else reading it knows I'm aware of what I'm doing. Now with this URL trick I can do both in one line. :)
Any time you want to sign up for any "open social" platform, you have to rush to claim your domain name. If someone else got there first, too bad. And that domain name applies to every app using this protocol. So no chance to claim it on another app, ever.
So what, exactly, is the difference between this and internet handles? In fact, isn't this worse?
3-in-1. Lack.
“The loudest voices preaching about taste and AI are often the ones who never demonstrated taste before AI.”
Yes, and if even these people can tell that AI generated stuff is godawful and tasteless, that tells you everything you need to know about AI.
The designer wants huge amounts of screen space wasted on unnnecessary padding, massive Fisher-Price rounded corners, and fancy fading and sliding animations that get in the way and slow things down. (Moreover, the designer just happens to want to completely re-design everything a few months later.)
The customer “ooh”s and “aah”s at said fancy animations running on the salesman’s top of the line macbook pro and is lured in, only realising too late that they’ve been bitten in the ass by the enormous amount of bloat that makes it run like a potato on any computer that costs less than four thousand dollars.
And US/EU laws are written by clueless bureaucrats whose most recent experience with technology is not even an electric typewriter.
What’s your point?
Be honest. How much of this article did you write, and how much did ChatGPT write?
This article could have been two lines. It takes some serious stretching of school-essay-writing muscles to inflate it to this many pages of waffle.
GPTBot et al. will probably do the same, as more people use AI to replace search.
It really won’t. It will steal your website’s content and regurgitate it back out in a mangled form to any lazy prompt that gets prodded into it. GPT bots are a perfect example of the parasites you speak of that have destroyed any possibility of an open web.
So just think how much greater the browser could be if Mozilla put more of the money they get into improving Firefox instead of into pointless UI redesigns that only slow things down, or breaking existing functionality - not to mention all the other frivolous nonsense they seem preoccupied with instead of being a credible competitor to Google.
With how they've been in recent years it's almost as if they're trying to be inept competition, as if they're being paid by Google to suck - in fact, that is all but established by now.
Let me introduce you to OpenCart, an open source eCommerce platform in use on hundreds of thousands of websites handling customer payments, recently struck a multi-million dollar deal with PayPal, and whose founder and practically sole developer responds to bug reports and CVEs with careful, well-thought-out replies like "JUST FUCK OFF!":
https://news.ycombinator.com/item?id=7526498 https://archive.is/9bHTi (archived version of github issue linked in above thread) https://www.theregister.com/2023/11/24/opencart_vulnerabilit... https://github.com/opencart/opencart/issues/12947#issuecomme...
I’m glad to see it. It really didn’t get much better than Desktop Cube back in the day.
Have they fixed 'Open in Conversation', the position of the New Message button, total number of messages indicator, tabs and toolbar buttons sharing the same row, the settings button, laggy UI, compatibility with various extensions or the search bar, all of which have been either completely broken or borderline unusable since their utterly unnecessary UI mangling in version 115, their self-proclaimed 'most beautiful Thunderbird yet'?
Or have they just added more padding around everything, more wasted screen space, more laggy animations, all the while making routine operations like, oh I don't know, checking and replying to email more unpleasant and niggling by the day?
I want to like this, but the page keeps reloading itself every few seconds. It’s really annoying.
For the rest of us, what does “echan” mean? Searching online for “irc echan” wasn’t much good.
edit: some wider-range searching suggests it might be Spanish for “kicked”.
A screen is not a piece of paper. Computers adopted white backgrounds with black text, when the technology became available, to imitate the paper and documents that people were most familiar with. But looking at a white screen for prolonged periods of time is like staring directly at a light bulb.
White text on a black background in print is indeed harder to read due to the issue of ink bleeding, but on a computer screen it is so much easier on the eyes.