• 0 Posts
  • 26 Comments
Joined 1 year ago
cake
Cake day: June 15th, 2025

help-circle


  • It’s a hammer being sold as a replacement for builders. I don’t hate AI, rather than all of stuff that’s called “AI” these days is being sold as something more than it is.

    Just listen to Jensen, the token salesman, of Nvidias fame.

    If that $500,000 engineer did not consume at least $250,000 worth of tokens, I am going to be deeply alarmed

    This is what I hate. The mad greed of these companies; the blatant disregard of everyone that these CEOs are displaying; the blind rush without a thought about consequences. “AI” is just the latest fad that got ruined by good old capitalism.


  • I don’t know how many times I had to deal with missing VCRUNTIME140.dll or MSVCP140.dll or other crap on Windows. This is not a Linux exclusive problem.
    Reading through the comment thread I can’t help but think that your whole situation is self imposed.

    Dependency problems are universal and there are tools to deal with it. It just seems that you’re refusing to use those tools (even Windows has winget now instead of relying on every installer bundling / linking its dependencies).
    Now, it’s fair to not want to deal with CLI, but your cited experience is an outlier. It is not normal to break your system with just apt update && apt upgrade -y. As a matter of fact apt will not upgrade if there are conflicting dependencies, you sort of have to force it to break your system.
    There are wrappers that provide a GUI for apt (and even dpkg, which is usually invoked when double clicking a .deb file) so why not using them?

    In Windows dependency issues are often offloaded to the provider of the software, but they are still just as present. In Linux this problem was solved[1] a different way — via package managers. I don’t want to be the “skill issue” guy, but refusing to use the platform intended tool to solve a problem is kind of a “skill issue”. At some point you are responsible for knowing how to use an OS, just as you are responsible for knowing how to drive a car if you want to drive a car.


    1. dependency hell is still an issue so take the word with a grain of salt. ↩︎









  • Mr. Satan@lemmy.ziptoLinux@lemmy.mlWhy?
    link
    fedilink
    arrow-up
    6
    ·
    11 months ago

    When Microsoft announced the sunset of Windows 10.

    I was still in uni at that time. Started with Ubuntu, disliked snaps and moved to Pop. Stayed there for last 5-ish (?) years. It does what I want it to do, I don’t care about switching distros now.





  • Well there you have it. Although I still feel weird that it’s somehow “the internet” that’s supposed to solve a problem that’s fully caused AI companies and their web crawlers.
    If a crawler keeps spamming and breaking a site I see it as nothing short of a DOS attack.

    Not to mention that robots.txt is completely voluntary and, as far as I know, mostly ignored by these companies. So then what makes you think that any them are acting in good faith?

    To me that is the core issue and why your position feels so outlandish. It’s like having a bully at school that constantly takes your lunch and your solution being: “Just bring them a lunch as well, maybe they’ll stop.”


  • Dunno, I feel you’re giving way too much credit to these companies.
    They have the resources. Why bother with a more proper solution when a single crawler solution works on all the sites they want?

    Is there even standardization for providing site dumps? If not, every site could require a custom software solution to use the dump. And I can guarantee you no one will bother with implementing any dump checking logic.

    If you have contrary examples I’d love to see some references or sources.