LLMs will FIX the internet, not ruin it; an argument:

Sam Altman recently acknowledged the large presence of LLMs on twitter and stated that he was finally starting to believe in dead internet theory.

And I couldn't agree more nor be any happier.

I had a sudden realization that social media has slowly crept from being "connect to your friends" to "try not to commit suicide as you climb the infinite social ladder amongst people who are not your peers"

... and I think it deserves to die. Furthermore, I think the only thing that could kill it is the greed that we're seeing finally strangle social media today.

People will never quit social media just because it's *bad for them*. It's too exciting, too interesting.

People will only quit social media if it itself becomes *bad*.

And when it becomes overbearingly full of corporate "brand personalities", LLM slop accounts churning out 500 posts a day to gain $4 in ad revenue, and 100 different AI bot accounts that add nothing to the conversation...

... that's when it's gonna be *bad* enough for people to reduce their usage of it. It will once again encourage the spaces we had before greed took a strangle-hold. Private areas for just you and your real-life friends. To be safe from the corporate slop bots, the LLMs, the 8-second videos of kittens performing impossible tasks... just you and your buds keeping up with each other.

This is the self-healing that nature is so good at. It's incredible how it always happens.

A programming language is a compromise between the absolute bullshit that a human wants to write and the absolute horsepiss that a computer wants to read.

Zig needs to get its shit together frankly.

I know the guy wants to perfect his little design, but has he never considered the fact that the shifting sands of perfection will never align for him?

Instead of working on making progress with the design he already established, he just keeps going back and changing fundamental parts as he sees fit and never making any forwards prog-

oh my god he's me.

!rant && coolThing

Today, I fully parallelized my evolution algorithm.

That means that since Thursday of last week, even though I took yesterday off, I was able to learn about metaheuristics, implement three of them (one being my own awful invention that did not perform) and successfully parallelize them.

Feeling good, man.

TL;DR: I'm reading papers and doing computer science like I could never afford to in college.

I am beginning my scientific arc.

Over the past few days, I have been working on implementing my own Evolutionary Algorithms

I've been doing a combination of "experimentation" and (probably less than I should,) actual research.

My Mark 1 was just a proof of concept that set up the data structures correctly, Mark 2 generalized the data structures and actually implemented some natural selection, but this was really just made up by me so I'm only getting mediocre results.

Next step: I have two papers lined up to read on EAs. Mark 3 might not implement them exactly, but I hope to beat the performance of Mark 2.

I'm encouraged by the fact that these research papers have TONS of different things they tried, and I'm really only on my first prototype (since Mark 1 didn't have any selection implementation, only randomness)

Follow along if interested:

https://github.com/AlgoRythm-Dylan/...

PSA: You setting up a RAG + System Prompt + Agent + Intents does NOT mean you “trained an AI”.

Happy Monday fellas. I'm gonna be pretty grumpy tomorrow morning because:

- Tonight's TO-DO list is writing unit tests and getting past a blocker to my project which I have been avoiding. Unit tests cause depression.
- After a recent bout of depression, I started smoking more weed than I was comfortable with and tonight starts my temporary sobriety. Withdrawal from THC causes more depression.

Wish me some God damned luck bros.

My secret addiction: defining all colors/ properties in CSS3 variables so that I can easily make my website theme-compatible, and then never making any themes for it.

Hey, at least it keeps my colors consistent....

@retoor account wiped and spamming comments, what's this about?

DOM custom events, and the Node.JS Event class are both so easy to work with and intuitive. Why doesn't JavaScript have native events? It was literally made for UI.

Also: I asked ChatGPT this exact thing before I posted because I had a sneaaaaking suspicion AND I WAS RIGHT. He called it a fuckin quirk.

ChatGPT is a 14 year old that just learned a new word and won't stop using it.

Anyone else unable to post comments on the app or webapp?

Trying to get flex container overflow to scroll is like funking pulling teeth. I've worked with flex containers for 5+ years straight and it still makes me want to kill myself.

Here's the tools you need to make a flex container scroll. God help you if it's nested more than 2-3 containers from the root. You apply these randomly until you see the scroll bar in the correct place.

min-height: 0px;
min-height: min-content;
overflow: auto;
overflow: hidden;
display: flex;
flex: 1;

(Real pro tip: you can place a "dummy" element inside the container you wish to make scroll. The dummy element should be a huge size, like 10k pixels. This allows you to see what containers up the DOM tree are still not correct (any container that is 10,000px is WRONG. Work your way down the DOM tree and fix fix fix))

If ChatGPT calls my problem a "classic" or it's solution a "subtle quirk of <framework>" one more fuckin time I'm gonna be the first human to murder an AI

I hate rollover video previews!!! Looking at you TouTube!!!! If I want a preview of the video, make it a button or something. The thumbnail disappearing and network activity lag every single time my mouse happens to barely caress an item in the video feed is SUPER FUCKING ANNOYING!

Especially on mobile!!!

Anyone else just randomly hit Help > About on programs when you’re bored?

In modern C#, everything's an interface.

Zig is actually sort of similar. I guess I owe it to C# for teaching me the magic.

Sorry, I mean:

IZig is IActually ISort of ISimilar. (IContinued...)

Wow, the new iOS design... looks like shit. It's hard to read text on top of the "liquid glass" and it's just sort of ugly looking.

How in the ever-god-fearing FUCK am I just now learning about git tag????

https://git-scm.com/book/en/...

Spent another 20 minutes blaming my tools when the issue was my fault again. Just wanted to let you friendly folks know.

(Forgot a “CREATE DATABASE IF NOT EXISTS” statement because on my other dev box, that database……. Already exists)

Just got WHIPLASH from seeing a racist ad in my local gym.

Racist is a bit dramatic, but at the VERY least, it's tone-deaf as fuck and from MY perspective (Where I missed the first 10 seconds of the ad) it was definitely racist feeling. Even with the context, I think it's pretty strange.

Anyways, this grocery chain is running this advertisement campaign about their yellow tags. I guess yellow tags in their store are supposed to be good! So they depict a bunch of yellow things and put text on screen saying that "yellow is (positive attribute etc etc)"

You might already see where this is going?

Basically, they had one scene with bold yellow text "YELLOW IS HAPPY" and the background of yellow tulips. Again, I joined the chat while the ad was already going so I barely had time to read all the text before the scene changed.

The next scene that they chose to depict yellow things?

Oh, just a child in yellow overalls holding a yellow ice pop. Oh, and the child was Asian...

SERIOUSLY how tone deaf could you POSSIBLY BE. DON'T PUT ASAINS IN YOUR COMMERCIAL ALL ABOUT YELLOW THINGS!!!!!!!!!!!

It was especially jarring since I lacked almost all context. I thought it was some shitty comedy sketch for a moment.

Link: https://youtube.com/watch/...

As mentioned in a previous rant, I'm not super happy about available graphical (s)ftp options.

Tried the command line and honestly, it's better than anything else.

The concepts are simple, and the tools are even better IMO. Being able to `put *` from a folder is great.

Why has nobody built an actually GOOD (s)ftp client yet? It's sort of crazy that Filezilla is STILL the first result on google (my AV correctly identified the installer as adware because it has by-default-enabled adware downloads) and then the next is windows-only.

I almost wish there was a nicely made Electron app (or - much better - qt) just so the UI could be fucking acceptable.

Is it time for me to start my next unfinished project??

I just got blocked on chess.com because some british guy blundered away like three pieces in a row and I went "blimey! the brexit geezer gambit!"

simple as.

If you have 100 hairs out of place nobody notices as it is just your style. However, if you have only one hair out of place everyone will notice it as it is a mistake.

Stack Overflow has the best April fools prank!

They are doing an 8+ hour database migration on a business day during business hours!

The prank is the fact that it’s not a joke! They really decided this for some reason!

Mojang/ Microsoft no longer understands semantic versioning??

They released a new update for Minecraft, 1.21.5 - but it:

- Adds new features
- Breaks backwards compatibility

Look at this. The left side of the screen was generated in 1.21.5, the right side was in 1.21.4 You can literally see the split down the middle where the trees generate differently because of the update.

This doesn't *matter* but it's just bothersome to me! Fuck you! I'll rant about whatever, no matter how small!

And after almost 4 months of trying, I have uninstalled the vim motions plugin for VSCode.

They're ok. I just never, I guess, understood them well enough to actually get a boost in performance.

Back to slinging my mouse around the editor, like God intended!

The VSCode material theme extension (extremely popular) has been found to be (potentially, but likely) malicious, and not only that, but the fucker that maintains it is a sleezy slimeball, too.

https://news.ycombinator.com/item/...

What's the mantra when there's no manual? RTFS?

Zig's standard library is largely RTFS. There's some docs, but it's a little sparse in my opinion.

Sharing progress on a project I started two days ago.

Elevator pitch: a personal (open-source) photography website where you can search my repository of photos by color, brightness, orientation, resolution, and subject.

I got a new camera and it actually performs very well and it made me realize that there's a lot of interesting data in images, so I've decided to convert my personal blog into a photography-centered personal blog.

My first idea, the one that actually drew me into starting the project, was to pick colors from my images (colorpicker/ eyedropper style) and then on the home page, those colors would appear in a grid and you could click them to see related photos.

Cool idea, but clunky and not very useful in practice.

So I implemented median-cut algorithm to generate a palette from an input image. It has its weaknesses, but it's consistent and does a job that makes you go "yeah that's about right" in 99% of cases. It generates 8 colors and then a second algo removes any colors that are within 5% distance from one another.

Before the color theory nerds ask, YES - I am using plain-old RGB and you can suck my balls. I don't care about human perception enough - not for this project at least.

Then, these colors (and other basic image analysis like brightness) are stored in the DB and related to the image entity from the upload.

Then the user (not yet implemented...) can do the reverse. Either they can choose colors from a colorpicker, or they can upload their own photo to get a palette and then the colors are looked up in the database and the related images are shown in the search results.

This will be combined with a somewhat complex tree-style system for the subject of the photo. For example, a subject like "crow" can be related to the subject "bird" which itself can be related to the parent subject "animal". So if you search by crow, you'll only get crows, but if you search "animal", you'll get birds, crows, etc.

While it's not exactly a reverse image search like google images has, I think it's a somewhat refined take on a way to explore a photography repository - especially a personal one.

I am developing the management portal and the public facing portals as separate projects due to separation of concerns - and so I can avoid implementing (and upkeeping) authentication for ONE user (me) so it just runs on my local network.

Screenshot is the current iteration of the upload page for my management portal, which will eventually take care of watermarks etc etc.