• 349 Posts
  • 1.29K Comments
Joined 1 year ago
cake
Cake day: April 4th, 2025

help-circle

  • Ich dachte, das soll der Markt regeln? Und der basiert doch auf “Survival of the fittest companies”, nicht wahr??

    Es wird übrigens auch immer betont, dass durch die ungesteuerte Transformation, die eben Insolvenzen verursacht, auch viele grosse Unternehmen pleite gehen.

    No Shit, Sherlock! In der geltenden Wirtschaftsphilosophie wachsen erfolgreiche Unternehmen fast immer, so viel sie können. Aber alte Unternehmen können sich nun mal nicht gut an neue Rahmenbedingungen anpassen. Dass alte Unternehmen eher an der Transformation weg von der fossilen Technik pleite gehen als junge, ist genauso eine Binsenweisheit wie dass Menschen die sterben, meistens älter sind, als welche die geboren werden. Ich persönlich kenn gar nicht viele Leute, die mit 100 Jahren und 1 Meter 75 gross geboren werden…



  • Die Kontrolleure beaufsichtigen auch, ob Unternehmen ihre gesetzlich vorgeschriebenen Eigenkontrollen korrekt durchführen. Dass diese Selbstkontrolle nicht immer funktioniert, zeigt ein Fall, der derzeit vor Gericht verhandelt wird.

    Vor dem Landgericht Kassel läuft seit Juli ein Prozess um einen der größten Lebensmittelskandale der vergangenen Jahre. Die Firma Wilke Waldecker Fleisch- und Wurstwaren soll zwischen 2015 und 2019 unter katastrophalen hygienischen Bedingungen Wurst hergestellt haben.

    Laut Staatsanwaltschaft war die Ware mit Listerien belastet, 37 Menschen sollen dadurch erkrankt, 11 gestorben sein. Der Vorwurf: Bei Wilke soll man von der Keimbelastung gewusst haben, die Ergebnisse der eigenen Untersuchungen aber über Jahre nicht an die Behörden gemeldet haben.

    Die nächste Evolutionsstufe dieses, äh, “industeriefreundlichen” Systems sehen wir gerade in den USA.














  • HaraldvonBlauzahn@feddit.orgOPtoLinux@lemmy.mlDebian and the sirens
    link
    fedilink
    arrow-up
    12
    arrow-down
    1
    ·
    edit-2
    6 days ago

    Unfortunately, one of the things that Debian is, is almost unable to manage packaging modern dependency trees. While more recent distributions like Guix can recursively import dependencies from a dozen programming languages’ package repositories, with a result that is generally acceptable to add to the distribution, Debian’s policies don’t make that very possible for a progam to accomplish. Perhaps some will use LLMs to do that. If they succeeed, Debian will become dependent on proprietary software for development, while still needing people in the loop, doing even less appealing scut-work.

    See also his blog entry on concrete disadvantages of contributions of LLM-generated code, in this case in git-annex.

    It is very interesting what Guix achieves.

    I run Debian on all my home systems, increasingly with some Guix on top.

    Yesterday, I was setting up an old X220 Thinkpad with Debian stable for a purpose of quick data entry in my living room. (My work Thinkpad can’t do that because due to some hardware problem, it crashes when one tries to suspend it and closes the lid). I wanted to have Taskwarrior on it, which I want to sync with the Taskwarrior instance on my Gemini PDA, for things like shopping lists. The PDA runs Sailfish OS, which contains Taskwarrior.Turns out it is Taskwarrior 3, which has a new sync protocol. Debian Trixie still has Taskwarrior 2 which is not compatible.

    The easiest solution? Installing Guix as package manager and “guix install taskwarrior”.

    And Guix also has a current Rust compiler and cargo build tool, which is nice for building and running personal tools,and things like jujutsu.

    Still, Debian is the base, and the contributions of Joey Hess and the other maintainers are invaluable.



  • And “AI” in general, and in a technical sense, is just a label for a set of techniques which have the common goal that computers solve certain, hitherto unsolved, complex tasks. The content of the set, and thus the exact meaning of the label changes over time. For example recognition of handwritten digits, graph algorithms e.g. for route planning, or backtracking algorithms playing board games were once considered AI, but aren’t today.

    Which is separate from the mere belief that computers will become as “intelligent” as humans - this is a kind of belief system with some quasi-religuous aspects.

    It would be better to not call it “AI” but “experimental novel computing methods”, leaving out the quasi-religuous stuff…






  • In the other corner, Proposal B seems to lay out something similar to how AI-generated code is handled on the Linux kernel. The proposal allows AI-generated code only if the submitter takes full accountability for it, ensures it’s free of copyright issues, and declares any LLM usage.

    Assuming full technical responsibility would probably make most AI-generated code pointless, as the whole point of using AI this way is that generating code saves work. Which is not any more the case if the result is really fully reviewed.

    This is different from using AI for vulnerability scanning (as it is practised in the Linux kernel).

    The other thing is there is no way one can declare AI generated code safely free of copyright isses. You’d need to prove that the result code was not contained in the training data - which is not possible.