this post was submitted on 30 Sep 2024
40 points (100.0% liked)

TechTakes

1432 readers
94 users here now

Big brain tech dude got yet another clueless take over at HackerNews etc? Here's the place to vent. Orange site, VC foolishness, all welcome.

This is not debate club. Unless it’s amusing debate.

For actually-good tech, you want our NotAwfulTech community

founded 1 year ago
MODERATORS
 

Need to let loose a primal scream without collecting footnotes first? Have a sneer percolating in your system but not enough time/energy to make a whole post about it? Go forth and be mid: Welcome to the Stubsack, your first port of call for learning fresh Awful you’ll near-instantly regret.

Any awful.systems sub may be subsneered in this subthread, techtakes or no.

If your sneer seems higher quality than you thought, feel free to cut’n’paste it into its own post — there’s no quota for posting and the bar really isn’t that high.

The post Xitter web has spawned soo many “esoteric” right wing freaks, but there’s no appropriate sneer-space for them. I’m talking redscare-ish, reality challenged “culture critics” who write about everything but understand nothing. I’m talking about reply-guys who make the same 6 tweets about the same 3 subjects. They’re inescapable at this point, yet I don’t see them mocked (as much as they should be)

Like, there was one dude a while back who insisted that women couldn’t be surgeons because they didn’t believe in the moon or in stars? I think each and every one of these guys is uniquely fucked up and if I can’t escape them, I would love to sneer at them.

Last week's thread

(Semi-obligatory thanks to @dgerard for starting this)

top 50 comments
sorted by: hot top controversial new old
[–] [email protected] 23 points 1 month ago* (last edited 1 month ago) (2 children)

snrk

https://mastodon.social/@Daojoan/113228461302575676

I’m going to start replying to everything like I’m on Hacker News. Unhappy with Congress? Why don’t you just start a new country and write a constitution and secede? It’s not that hard once you know how. Actually, I wrote a microstate in a weekend using Rust.

[–] [email protected] 9 points 1 month ago

Actually, I wrote a microstate in a weekend using Rust.

I'm dead. At least the Rust Evangelism Strike Force finally got to have their theocracy

load more comments (1 replies)
[–] [email protected] 20 points 1 month ago (3 children)

I wanted to see how much lactose was in Monterey Jack, and this was the very first result on bing for:

monterey jack lactose per 10 grams

https://thekitchencommunity.org/the-nutritional-profile-of-monterey-jack-cheese/

It's absolutely over. This is why every other search I make has "site:reddit.com" attached to it.

And no, the site didn't tell me how much lactose there was per gram

[–] [email protected] 11 points 1 month ago (4 children)

I can’t decide if calimisinit is better pronounced in a surfer bro or British accent, so my brain combined the two and I hate it

load more comments (4 replies)
[–] [email protected] 10 points 1 month ago

"My name is Scroder Cher. I take care of the place while the Master is away."

load more comments (1 replies)
[–] [email protected] 18 points 1 month ago (8 children)

Found while poking around today: the Wikipedia club for cleaning up after AI.

Example: the article Leninist historiography was entirely written by AI and previously included a list of completely fake sources in Russian and Hungarian at the bottom of the page.

load more comments (8 replies)
[–] [email protected] 18 points 1 month ago (1 children)

I’m just thinking about all the reply guys that come here defending autoplag, specifically with this idea:

“GPT is great when I want to turn a list of bullet points into an eloquent email”

Hey, you butts, just send the bullet points! What are you, a high schooler? Nobody has time for essays, much less autoplagged slop.

[–] [email protected] 15 points 1 month ago

No no no it's fine! You get the word shuffler to deshuffle the—eloquently—shuffled paragraphs back into nice and tidy bullet points. And I have an idea! You could get an LLM to add metadata to the email to preserve the original bullet points, so the recipient LLM has extra interpolation room to choose to ignore the original list, but keep the—much more correct and eloquent, and with much better emphasis—hallucinated ones.

[–] [email protected] 15 points 1 month ago (1 children)

a user called "crypto-bitcoin" raises an issue with the World Wide Web Consortium's Accessibility List

[–] [email protected] 11 points 1 month ago* (last edited 1 month ago) (1 children)

See image description below

Image descriptionImage shows user joined two weeks ago.
.
Yikes. Could be a troll (I hope it's a troll)

[–] [email protected] 12 points 1 month ago

a lost not very computer literate dude who just got scammed i guess

[–] [email protected] 14 points 1 month ago (1 children)

ah, sweet, manmade horrors beyond my comprehension:

manmade horrors

(This would've been more shocking to me in 2023, but after over a year in this bubble I have stopped expecting anything resembling basic human decency from those who work in AI)

load more comments (1 replies)
[–] [email protected] 14 points 1 month ago (7 children)

Today in "Promptfondler fucks around and finds out."

So I'm guessing what happened here is that the statistically average terminal session doesn't end after opening an SSH connection, and the LLM doesn't actually understand what it's doing or when to stop, especially when it's being promoted with the output of whatever it last commanded.

Shlegeris said he uses his AI agent all the time for basic system administration tasks that he doesn't remember how to do on his own, such as installing certain bits of software and configuring security settings.

Emphasis added.

[–] [email protected] 12 points 1 month ago

"I only had this problem because I was very reckless," he continued, "partially because I think it's interesting to explore the potential downsides of this type of automation. If I had given better instructions to my agent, e.g. telling it 'when you've finished the task you were assigned, stop taking actions,' I wouldn't have had this problem.

just instruct it "be sentient" and you're good, why don't these tech CEOs undersand the full potential of this limitless technology?

load more comments (6 replies)
[–] [email protected] 14 points 1 month ago (12 children)

So, today MS publishes this blog post about something with AI. It starts with "We’re living through a technological paradigm shift."... and right there I didn't bother reading the rest of it because I don't want to expose my brain to it further.

But what I found funny is that also today, there's this news: https://www.theverge.com/2024/10/1/24259369/microsoft-hololens-2-discontinuation-support

So Hololens is discontinued... you know... AR... the last supposedly big paradigm shift that was supposedly going to change everything.

load more comments (12 replies)
[–] [email protected] 13 points 1 month ago* (last edited 1 month ago) (6 children)

Small FYI, not a sneer or anything, you can stop reading if you don't know what the godotengine is. But if you do and hear of the fork, you can just ignore the fork. (the people involved also seem to be rather iffy, one guy who went crazy after somebody mentioned they would like gay relationships in his game, and some maga conspiracy theory style coder. That is going by the 3 normal people the account follows (out of 5) who I assume are behind it).

[–] [email protected] 21 points 1 month ago* (last edited 1 month ago) (28 children)

amazing work they've achieved so far though

yes, that's the complete change set for the fork

oh and a Godot discord moderator turned out to be extremely into dropping n-words in chat

[–] [email protected] 14 points 1 month ago

This is fucking hilarious. GIT is my favorite blockchain

load more comments (27 replies)
[–] [email protected] 13 points 1 month ago (1 children)

There's a new fork of Godot!

It's called Lolidot, because of course it is

it's going great too

[–] [email protected] 10 points 1 month ago

Of course it could get worse.

Lol at them just removing the godot donation links/list of donors and just replacing it with their own. What a grift.

[–] [email protected] 11 points 1 month ago

god these people are fucking losers

load more comments (3 replies)
[–] [email protected] 13 points 1 month ago (3 children)
load more comments (3 replies)
[–] [email protected] 13 points 1 month ago (1 children)
load more comments (1 replies)
[–] [email protected] 13 points 1 month ago* (last edited 1 month ago) (9 children)
load more comments (9 replies)
[–] [email protected] 13 points 1 month ago (2 children)
load more comments (2 replies)
[–] [email protected] 11 points 1 month ago

so, I've always thought that blind's "we'll verify your presence by sending you shit on your corp mail" (which, y'know, mail logs etc....) is kinda a fucking awful idea. but!

this is remarkably fucking unhinged:

[–] [email protected] 11 points 1 month ago (2 children)
[–] [email protected] 13 points 1 month ago* (last edited 1 month ago) (1 children)

Elon musk voice: 'So euhmmm euuhhh ummm you are ummm euhh saying it has umm some value?'

E: But wait there is more

Tweet from Paul Graham: "Renaming Twitter X doesn't seem to have damaged it. But it doesn't seem to have helped it either. So it was a waste of time and a domain name."

And more

Reply tweet from Elon Musk: "You know nothing"

(And as is tradition when Musk replies to a famous person, the famous person doesn't even react).

[–] [email protected] 10 points 1 month ago (1 children)

The tragedy of being Elon Musk: he has thousands and thousands of cryptocurrency bros and fascists that worship the ground he walks on, but what he really want is for Stephen King to reply to one of his tweets.

[–] [email protected] 10 points 1 month ago (7 children)

Stephen King is busy working him in as a villain in a new novel.

load more comments (7 replies)
[–] [email protected] 11 points 1 month ago (1 children)

This gonna end in a gofundme

[–] [email protected] 14 points 1 month ago
[–] [email protected] 9 points 1 month ago* (last edited 1 month ago) (9 children)

I might be wrong but this sounds like a quick way to make the web worse by putting a huge computational load on your machine for the purpose of privacy inside customer service chat bots that nobody wants. Please correct me if I’m wrong

WebLLM is a high-performance in-browser LLM inference engine that brings language model inference directly onto web browsers with hardware acceleration. Everything runs inside the browser with no server support and is accelerated with WebGPU.

WebLLM is fully compatible with OpenAI API. That is, you can use the same OpenAI API on any open source models locally, with functionalities including streaming, JSON-mode, function-calling (WIP), etc.

We can bring a lot of fun opportunities to build AI assistants for everyone and enable privacy while enjoying GPU acceleration.

https://github.com/mlc-ai/web-llm

load more comments (9 replies)
load more comments
view more: next ›