Mastodon Feed: Posts

dysfun@treehouse.systems ("gaytabase") wrote:

i don't know how to put this more simply:

• fascism is bad
• reporting fascism is not
• fascists should have the shit kicked out of them frankly, but we'll settle for banning them
• people who report fascists for fascist comments should not be banned.

if you can't manage that, you have no business calling yourself 'community', 'inclusive' or any of the other nice words you like to attribute to yourselves.

dysfun@treehouse.systems ("gaytabase") wrote:

i swear i lose brain cells every time i read kde people this week

Boosted by soatok@furry.engineer ("Soatok Dreamseeker"):
UnicornRiot ("Unicorn Riot") wrote:

NEW: A Failed Reception System in Crete Turns People on the Move Into Political Targets

Residents of a small village in Northern Crete are fighting the Greek government’s plan to convert a disused factory into a reception center for people on the move, escalating a monthslong standoff that pits xenophobic backlash against an increasingly securitized European border regime.

https://unicornriot.ninja/2026/a-failed-reception-system-in-crete-turns-people-on-the-move-into-political-targets/

ireneista@irenes.space ("Irenes (many)") wrote:

we're thinking we want to maybe do incremental stable releases about every month or so, so right now this is functionality you can get in the main branch but not at the release tag

ireneista@irenes.space ("Irenes (many)") wrote:

we kept it as cat instead of an input redirection, in the README, because we want to subtly call attention to the fact you can use it to combine multiple source files. that said, you no longer need to :)

ireneista@irenes.space ("Irenes (many)") wrote:

we would share the original command to compile it under the hex transform, but that wouldn't fit in a toot

the new one is

$ cat inscribe-evoke.e | ./evoke > evoke.hex

a significant improvement

ireneista@irenes.space ("Irenes (many)") wrote:

and here is the command to compile it now

$ cat evoke.e | ./evoke > evoke2

ireneista@irenes.space ("Irenes (many)") wrote:

here is the the command to compile Evocation as it was documented when we did the release

$ (cat labels.e elf.e transform.e execution.e; echo 's" pyrzqxgl" allocate-string dup 262144 read-to-buffer'; cat core.e linux.e output.e amd64.e execution-support.e log-load.e; echo pyrzqxgl swap 262144 read-to-buffer; cat core.e linux.e output.e amd64.e execution-support.e log-load.e dynamic.e input.e interpret.e flow-control.e linux-dynamic.e ; echo pyrzqxgl; cat evoke.e) | ./evoke > evoke2

ireneista@irenes.space ("Irenes (many)") wrote:

yesssssss. it does work. we just checked it in (it'll appear on our site in a minute or so, there's a sync process that mediates between our personal code-review tool, in which we review our own code before approving it because we're like that, and our code browsing site. the latter is the public one.)

Boosted by soatok@furry.engineer ("Soatok Dreamseeker"):
geist@xoxo.zone wrote:

@danluu Some of them will end up being real, and then the click-worthy story is “Anthropic’s AI found in , along with 93 others”

Boosted by soatok@furry.engineer ("Soatok Dreamseeker"):
geist@xoxo.zone wrote:

@danluu I feel like there’s a straightforward answer, which is that it wasn’t in their interest to care about false positives. Anthropic, as an organization, doesn’t care about bugs in arbitrary open source projects. The reports are a form of marketing, for which large numbers are better.

Boosted by soatok@furry.engineer ("Soatok Dreamseeker"):
danluu ("Dan Luu") wrote:

Why didn't Anthropic do effective false positive rejection with their Mythos/Glasswing vuln reports? In https://danluu.com/ai-coding/#mythos, I mentioned a colleague finding that the reports we got were mostly nonsense.

At the time, I wondered our reports were particularly bad, but as more people talk about their experiences, getting mostly nonsense reports seems to have been common? E.g., here's @gregkh on Linux kernel reports (h/t @KernelRecipes); see also https://www.vulncheck.com/blog/anthropic-glasswing-receipts.

![] »7 — “assume a malicious filesystem image” »2 - “assume you can inject a malicious network packet into the middle of the stack” »2-NOMMU » 6 — sctp networking issues for untrusted devices »2 — minor ipv6 network issues HL : — gpu driver issue for malicious local user diy ]7

Boosted by soatok@furry.engineer ("Soatok Dreamseeker"):
thisness@beige.party ("Flux") wrote:

There is a strange thing that happens when you have spent most of your life surviving. You become very good at it. You learn to read the room before anyone speaks. You learn which footsteps mean trouble. You learn to anticipate moods, swallow needs, stay useful, stay busy, stay alert. You become capable. Independent. Productive. Maybe even impressive. You can carry an enormous amount because you learned early that somebody had to. And then, sometimes years later, you finally reach a place where the war is over. And you are exhausted.

— God’s Huntress

dysfun@treehouse.systems ("gaytabase") wrote:

kde's determined to lose, aren't they?

ireneista@irenes.space ("Irenes (many)") wrote:

hmmm we think we've figured out how to make include work under the hex transform, so that solves our immediate problem

we're not clear that making it work under the label or log-load transforms is viable (it's unclear what the semantics should be), but maybe it doesn't need to. hmmmmm

Boosted by baldur@toot.cafe ("Baldur Bjarnason"):
uglyreykjavik.bsky.social@bsky.brid.gy ("Ugly Reykjavik") wrote:

The vegetation next to the house is so thick I couldn't get through it to take a photo.

#Iceland #Reykjavik #photography #streetphotography #naturephotography #nature #abandoned #decay #windows

Boosted by baldur@toot.cafe ("Baldur Bjarnason"):
mhoye@cosocial.ca wrote:

These days my first reactions to projects talking about AI in open source aren't even about the ethical disaster, the environmental disaster, the theft and gating of the commons or an of the other ludicrous externalities this blight of an industry has inflicted on all of us.

My first, most visceral reaction to people who should know better and spent years claiming they did is, what happened to your self-respect. Did you ever bother trying to understand the things you said you believed in?

Boosted by baldur@toot.cafe ("Baldur Bjarnason"):
timnitGebru@dair-community.social ("Timnit Gebru (she/her).") wrote:

By me and @emilymbender.

"According to the AI industry, we should be more worried about a fictional machine god than about the climate catastrophe these data centers exacerbate, the asthma suffered by those living near them, the rising electricity bills of the public subsidizing them, or the water...[for] cooling them."

https://www.technologyreview.com/2026/09/22/1144867/dont-be-fooled-summer-ai-hype/

Boosted by ireneista@irenes.space ("Irenes (many)"):
msbellows@c.im ("M.S. Bellows, Jr.") wrote:

I'm so old, I remember when software that spread into the wild and infiltrated other people's computers and did harmful things to them that the users didn't ask it to was called "malware" instead of "AI" and the perpetrators were prosecuted instead of given billions in venture capital and government contracts.

Boosted by cwebber@social.coop ("Christine Lemmer-Webber"):
wehpudicabok@kind.social ("flower bathed in light") wrote:

cwebber@social.coop ("Christine Lemmer-Webber") wrote:

stinkpad

Boosted by cwebber@social.coop ("Christine Lemmer-Webber"):
mcc wrote:

@elebertus The machine is designed to fool people. It's designed to present a sentence, image, or block of code that makes a human think it is looking at a sentence, image, or block of code. It's is designed to lie *and* designed to make you trust it. Therefore there is no such thing as a person who can use it "well", or review its outputs well. There is only a person who is not yet far enough down the trained-to-be-jaded-and-just-press-the-button curve to let major problems slip past them *yet*

Boosted by cwebber@social.coop ("Christine Lemmer-Webber"):
dthompson@toot.cat ("dave") wrote:

and @ridley is making a completely independent implementation! this isn't a spritely walled garden.

Boosted by cwebber@social.coop ("Christine Lemmer-Webber"):
dthompson@toot.cat ("dave") wrote:

someone should give spritely one of those 3 million dollar donations because we're about to have an end-to-end encrypted relay network that will enable self-hosting all sorts of shit at home or wherever and it's gonna be time to start building little apps. imagine a fediverse where your identity and data are both decoupled from where people send messages to you.

cwebber@social.coop ("Christine Lemmer-Webber") wrote:

Open source procedural material authoring tool, made in Godot https://www.materialmaker.org/

Seems damn cool

chris@fedi.freebooters.uk ("Chris Were ⁂🐧🌱☕️") wrote:

We all suck at Garry's Mod
https://chriswere.wales/videos/media/projectchronicle/2014-01-07%20We%20all%20suck%20at%20Garry's%20Mod.webm

jonny@neuromatch.social ("jonny (nonvenomous)") wrote:

The transparency about the system is becoming a bit clearer, there is a skill in here about self-knowledge and it does say to just be honest and show people the computer. There is a whole part of the app that lets you directly browse the /home folder, including the ssh keys, which is a very strange choice. But this doesn't have the skills and the more fun stuff.

Boosted by cwebber@social.coop ("Christine Lemmer-Webber"):
SwiftOnSecurity@infosec.exchange wrote:

Artists realizing alt-text trained AI on their art style

Boosted by pzmyers@freethought.online ("pzmyers 🕷"):
gildilinie@beige.party ("Gildilinie Gremlin 🏳️‍⚧️🍳") wrote:

@davidgerard look. in 2023 i was SURE that people who knew the guts of the chatbot like me would never fall for this shit and I was SO WRONG

jonny@neuromatch.social ("jonny (nonvenomous)") wrote:

I'm going to give it a temporary credit card with $20 on it and tell it to go out and fine me some fine wares, I'm not going to give it any real creds, but I am going to try and construct a sort of Potemkin social graph to see what it says. I am using another LLM to make a fake social media site with an open api and populate it with demo users to see how the social surveillance markdown works. I am going to try to a) get it to think I am in love with someone who loves to buy a specific category of product to see how the shopping recs transfer, and b) get it to think I am part of some secret conspiracy and see if it tries to root it out. Taking recs for scenarios to test the surveillance.