jannic

@bagder@mastodon.social Enjoy the summer!

@technomancy @glyph Didn't that happen decades ago, long before LLMs?

@argv_minus_one@mastodon.sdf.org @recursive@hachyderm.io Some people build incredible things in their garage or shed: https://www.youtube.com/watch?v=h6GWikWlAQA

@fetopa Dabei darf man nicht übersehen: Wenn die gesetzliche Rente nicht über dem Existenzminimum liegt, dann zahlt man in die Rentenversicherung nicht ein, weil man später etwas davon hat, sondern nur, damit später der Staat nicht die Kosten hat um das Existenzminimum zu sichern. Der Begriff "Versicherung" ist damit falsch. Es ist eine Steuer.

@stanfromireland@mastodon.social @bagder@mastodon.social I wonder if Daniel's complaints became popular enough, that the AIs now give curl some special treatment. Just to avoid bad publicity. (But then, if the vendors were able to tune the AIs so they don't send slop reports to curl, why should they limit that treatment to curl? Does it consume significantly more resources to detect invalid reports?)

@helenasteinhaus ich fände total überhöhte Abgeordnetendiäten vollkommen in Ordnung, wenn die dadurch immun für jede Form von Korruption würden. Leider funktioniert das so nicht. Mehr Geld macht anscheinend nur noch gieriger. 😢

@kpcyrd @tekkie And the vulnerability is really narrow: To be affected, you need to download a malicious crate. Note that as soon as you compile such a malicious crate, it can do harm much more directly because compiling rust code inherently allows for arbitrary code execution through build.rs and proc macros.

@bunnie I love how a well written and interesting blog post attracts high quality comments. Even with proper citations! 😍 https://www.bunniestudios.com/blog/2026/baochip-1x-a-mostly-open-22nm-soc-for-high-assurance-applications/#comment-2754163

@jann@infosec.exchange These chips do have an MPU (memory protection unit, like an MMU but without the capability to remap memory addresses). So MSPLIM is not strictly necessary to implement some kind of stack overflow protection, but much easier to set up. Additional hardware overhead could be a reason it's not used on bigger CPUs, yes. And with many cores, each core would need a separate stack limit.

@jann@infosec.exchange @ljrk@todon.eu I wonder why it is still necessary to employ imperfect mechanisms like guard pages to avoid stack overflows. Shouldn't it be possible to have a stack pointer limit checked by hardware? Like https://interrupt.memfault.com/blog/using-psp-msp-limit-registers-for-stack-overflow on some ARM chips?

@donnerbella@mastodon.social Wirtschaftlicher Erfolg ist kein Selbstzweck. Wirtschaft muss zu nachhaltiger Verbesserung der Lebensqualität aller Menschen führen.

@jimbob@aus.social While this might work out for the corporations right now, I wonder if they have a plan for the longer term. I mean: LLMs might solve some coding problems, and may even replace a lot of junior developers in the short term. (I doubt even that, but let's just assume it for now.) But you will still need experienced people for problems LLMs can't solve. And these people don't grow on trees, they were inexperienced juniors before. If you break that career path, you'll end up with LLMs for the easy problems and no solution at all for more difficult ones.