Mastodon Feed: Posts

soatok@furry.engineer ("Soatok Dreamseeker 🔜 Megaplex") wrote:

About to suit up and head down

#megaplex2026

isagalaev ("Ivan Sagalaev :flag_wbw:") wrote:

RE: https://infosec.exchange/@david%5Fchisnall/117134446424182178

Something I've been telling people from the beginning of this bubble: LLMs are showing us where we need better libraries.

Boosted by jsonstein@masto.deoan.org ("Jeff Sonstein"):
dgoosens@phpc.social ("Dmitri Goosens :elephpant:") wrote:

OMG... this is so spot on

Boosted by jsonstein@masto.deoan.org ("Jeff Sonstein"):
heidilifeldman ("Heidi Li Feldman") wrote:

Spot on.
“You can’t have a Justice Department or FBI doing corrupt work with one hand and normal aboveboard work with the other and expect anyone — from citizens to lawyers to judges — to decipher which is which in any given case.”
https://talkingpointsmemo.com/morning-memo/how-trumps-corruption-makes-everything-doj-does-suspect

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

RE: https://autonomous.zone/@p2panda/117134466795114150

Horray, two friendly peer projects (at Spritely we see Willow and p2panda both as allied projects) working together!

And the announcement video is a ton of fun, nice work @gwil !

andrewnez ("Andrew Nesbitt") wrote:

That feeling when you see a company piss someone off so much that they change their license to APGL 😂

Boosted by cwebber@social.coop ("Christine Lemmer-Webber"):
p2panda@autonomous.zone wrote:

Willow + p2panda = ..?

Can this work? .. 🙀 They said it couldn't be done? Find out in this video:

https://spectra.video/w/3Dncy2G13XyfEEFVteipcY

Very excited to annouce this wonderful collaboration with our friends of Worm Blossom as part of the @nlnet NGI Zero Commons Fund! 😻

More in their blog with a little guest feature of what the pandas have been listening to lately:

https://worm-blossom.org/#y2026w34

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

@cwebber Once you break the barrier to letting the models approve reviews of model code and merge without human oversight, it seems difficult to find another red line that will hold, to say nothing of reversing the trend. That is, until a catastrophe makes that decision for you.

To clarify, that hasn't happened yet, but it's one step away. Or, it will happen and nobody will admit to it, because the human reviewers are actually shunting the work to models.

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

Yes I missed this line:

> Quick and dirty count suggests we (Paolo and I) merged a very similar number of net (632) and net-next (648) patches. This is not telling the full story either because 1/3 to 1/2 of the net-next patches also *seem* like AI-driven low priority fixes, cleanups and clarifications.

Now would be a STELLAR time if you wanted to introduce vibe-obscured malware into the Linux kernel. (I forget where I read the research on it, but LLMs have been shown to be pretty good at this, inserting intentional vulnerabilities in ways that are hard to detect, embedded in other patches.)

Boosted by cwebber@social.coop ("Christine Lemmer-Webber"):
AndrewRadev@hachyderm.io ("Andrew Radev") wrote:

In case you were wondering how the linux kernel is doing: https://lore.kernel.org/lkml/20260818212923.2725202-1-kuba@kernel.org/

> This is not telling the full story either because 1/3 to 1/2 of the net-next patches also *seem* like AI-driven low priority fixes, cleanups and clarifications.
> We are completely overwhelmed, of course. The glimmer of hope is that we secured sufficient LLM budget and access (thank you Meta!) to run reviews with multiple frontier models on each patch. This eliminates some hallucinations.

That's right, "Thank you Meta!", this kernel maintainer enthusiastically exclaims. Presumably, for subsidizing their slop machines trained on pirated books with money from online fraud. Which eliminates "some" hallucinations. An incredible achievement, I'm sure.

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

thank you Meta!

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

From the Linux Kernel Mailing List: https://lore.kernel.org/lkml/20260818212923.2725202-1-kuba@kernel.org/

> We are completely overwhelmed, of course. The glimmer of hope is that we secured sufficient LLM budget and access (thank you Meta!) to run reviews with multiple frontier models on each patch. This eliminates some hallucinations. That said, in terms of review, the LLMs can only do so much. The sad truth is that our APIs (especially for rare events like PCIe errors, timeouts etc.) have always been racy, and now LLMs don't let us ignore that. I expect our direction for the next release will be to tweak the reviews a little bit more, but start shifting focus to letting the LLMs take care of the busy work - managing patchwork, automating common process complaints, editing commit messages, and maybe applying patches which already got "reviewed-by" tags from people we trust...

I look at the above and can't help but wonder if this is the "vibe bobsled" in action https://dustycloud.org/blog/faulty-towers-vibe-sickness-and-the-vibe-bobsled/

NfNitLoop ("Cody Casterline 🏳️‍🌈") wrote:

@musicmatze ah yeah. I need to look into licenses and make a choice there.

Boosted by jsonstein@masto.deoan.org ("Jeff Sonstein"):
markarayner@mas.to ("Mark A. Rayner") wrote:

In all fairness, there should be a sign or something.

#dog #dogs #dogsofmastodon #humor #humour

jsonstein@masto.deoan.org ("Jeff Sonstein") wrote:

JFC, President Bonespurs has absolutely no shame

https://www.theguardian.com/us-news/2026/aug/21/navy-carrier-black-war-hero-trump-report

Boosted by cstanhope@social.coop ("Your weary 'net denizen"):
grickle@mstdn.social ("Grickle") wrote:

Endless. #grickledoodle #sloth #scrolling #phones #cartoon #rage #art #drawing #funny #humor #sad

Boosted by jsonstein@masto.deoan.org ("Jeff Sonstein"):
paninid@mastodon.world ("Coach Sankhavaram ®") wrote:

“Never start a land war in Asia”

- #Herodotus

#HistoryMemes #history

andrewnez ("Andrew Nesbitt") wrote:

This is giving me a lot of crypto-testnet vibes: https://atproto.com/blog/atproto-spaces-alpha

jsonstein@masto.deoan.org ("Jeff Sonstein") wrote:

lunch in claw

andrewnez ("Andrew Nesbitt") wrote:

Two-Factor Authentication Across Package Registries

https://nesbitt.io/2026/08/18/two-factor-authentication-across-package-registries.html

Boosted by jsonstein@masto.deoan.org ("Jeff Sonstein"):
iftas@mastodon.iftas.org ("IFTAS") wrote:

The Russian botnet has now created over 2,500 inauthentic accounts on 180 Mastodon servers. These accounts spam propaganda, follow each other to increase federation, connect to Bluesky via the bridge to spread their content there, and create unlimited invite links for their bots to follow.

IFTAS reports each one we find, and to date roughly 60% of these accounts have been suspended.

Our AUD denylist would silence or suspend the bulk of the remaining 40% of accounts.

https://about.iftas.org/library/suspected-portal-kombat-accounts/

Boosted by andrewnez ("Andrew Nesbitt"):
reproducible_builds@fosstodon.org ("Reproducible Builds") wrote:

diffoscope 329 just released 💠 https://diffoscope.org/news/diffoscope-329-released/ by @lamby

jsonstein@masto.deoan.org ("Jeff Sonstein") wrote:

oh, okay. thanks for the warm greetings, @CARROT

Boosted by neatnik@social.lol ("Neatnik :prami:"):
jpoesen@social.jpoesen.com ("jpoesen | 🇪🇺 | 🏳️‍🌈") wrote:

Following a discussion on #IndieWeb initiated by @neatnik I've published my thoughts on Softcore IndieWeb and #Smallscale #Drupal.

https://jpoesen.com/articles/2026/08/21/on-softcore-indieweb-and-smallscale-drupal/

jscalzi@threads.net ("John Scalzi") wrote:

They need to get a good look at the bus they're pushing trans people under, the GOP is going to shove them under those same wheels the next time it loops around

https://www.lgbtqnation.com/2026/08/log-cabin-republicans-announce-they-no-longer-support-trans-rights/

Boosted by brib@bribstodon.xyz ("brib :neofox_floof:​ :Nonbinary:"):
o76923@kitty.social ("James Endicott") wrote:

You know that Amnesty International report calling JKR anti-rights? Here's the magnet link to it, since she's issuing takedowns to have it removed from Google Drive

magnet:?xt=urn:btih:eb3e6f89f9c71404643bc7b9bb6d74963017dda9&dn=Report_A_growing_threat_the_anti_rights_movement_in_the_UK_July.pdf

andrewnez ("Andrew Nesbitt") wrote:

I think this is the cutest car I've ever owned.

Boosted by glyph ("Glyph"):
SwiftOnSecurity@infosec.exchange wrote:

RE: https://mastodon.social/@glyph/117119121706515076

jscalzi@threads.net ("John Scalzi") wrote:

Oooof and also yes

RE: https://www.threads.com/@tananarivedue/post/DcTp4%5FrGzCo

Boosted by baldur@toot.cafe ("Baldur Bjarnason"):
emilymbender@dair-community.social ("Prof. Emily M. Bender(she/her)") wrote:

Because it only works because the readers can do the same thing, and construct what's going on for the writer by imagining their point of view.

When we do this with other people, it is lovely. Literally the stuff of relationships, communities, societies.

But it is also fundamental to how we interpret language and we can't just turn it off. And that is what chatbot purveyors are preying on with their product design.

/fin