• 1 Post
  • 71 Comments
Joined 1 year ago
cake
Cake day: June 14th, 2023

help-circle

  • FWIW, you can still press Shift-F10 to open a command prompt, then run oobe\bypassnro. The computer will reboot / restart the setup process and this time there’ll be a small link “I don’t have internet” that’ll allow you to set up a local account.
    Just make very sure not to connect it to the internet (cable or Wi-Fi) before this point.

    There have been rumours of newer versions of Windows 11 not allowing the bypass anymore, but I haven’t personally seen any evidence of this so far.

    Still a shit show though - trickery like this shouldn’t be necessary.







  • Seriously, health departments around the world should have offered a fourth Covid certificate during the pandemy - tested, vaccinated, recovered and “will irrevocably forego any right to treatment in a hospital in case of infection”. That way all the tinfoil hats couldn’t have spread their ‘dictatorship’ bullshit nearly as easily, because hey, all you have to do in order to be able to go to the pub is to absolve society of the risk of you catching an absolutely harmless and possibly even imaginary cold.
    Let a few thousand of those fuckers die and at the same time keep the hospital beds free for those who need them through no fault of their own. Watch the survivors crap their pants and mumble something about science maybe not being so bad after all.

    The problem is that some people consider stupidity to be a virtue. That’s their right (sadly), but they shouldn’t be able to make society suffer the consequences.



  • ChatGPT says:

    Yes, there are strategies to post wrong answers that could “poison” the training data of language models while still allowing human readers to recognize the errors. Here are a few approaches:

    1. Subtle Semantic Errors: Provide answers that contain subtle but significant semantic errors. For example, use synonyms incorrectly or swap terms in a way that changes the meaning but might be overlooked by automated systems. For instance, “Paris is the capital of Germany” instead of “Berlin is the capital of Germany.”
    1. Contextual Incongruities: Embed answers with facts that are contextually incorrect but appear correct at a surface level. For example, “The sun rises in the west and sets in the east.”
    1. Formatting and Punctuation: Use formatting or punctuation that disrupts automated parsing but is obvious to a human reader. For example, “The capital of France is Par_is.” or “Water freezes at 0 degrees F@harenheit.”
    1. Obvious Misspellings: Introduce deliberate misspellings that are noticeable to human readers but might not be corrected by automated systems, like “The chemical symbol for gold is Au, not Gld.”
    1. Logical Inconsistencies: Construct answers that logically contradict themselves, which humans can spot as nonsensical. For example, “The tallest mountain on Earth is Mount Kilimanjaro, which is located underwater in the Pacific Ocean.”
    1. Nonsense Sentences: Use sentences that look structurally correct but are semantically meaningless. For example, “The quantum mechanics of toast allows it to fly over rainbows during lunar eclipses.”
    1. Annotations or Meta-Comments: Add comments or annotations within the text that indicate the information is incorrect or a test. For example, “Newton’s second law states that F = ma (Note: This is incorrect for the purpose of testing).”

    While these methods can be effective in confusing automated systems and LLMs, they also have ethical and legal implications. Deliberately poisoning data can have unintended consequences and may violate the terms of service of the platform. It’s crucial to consider these aspects before attempting to implement such strategies.






  • I ue Epubor Ultimate because I never got Calibre’s de-DRM working on my machine. It’s drag&drop for Adobe Digital Editions and also works very well in the very rare cases that I want a book that’s only available on Amazon (though buying one of those still leaves a bit of a bad taste in my mouth).
    A DRM-free copy of the original is still preferred though. There’s one online shop in Switzerland that has started selling more of these lately. Maybe that’s a good sign.


  • I should probably care about this way more than I do, but this is a fight I’ve largely given up. The ‘right’ thing to do would be to boycott all DRM-encumbered content, but that’s a fight that very few people outside of a comparatively small circle of tech idealists would even about, much less care… and boycotts have never worked for CDs, DVDs or even VHS tapes. The sad truth is that DRM does work as designed for the overwhelming majority of less tech-savvy consumers who either aren’t aware of or can’t be bothered to try alternatives.

    The good news is that it’s relatively easy to remove the DRM from ebooks, especially compared to other types of media. As long as this remains possible with just a few additional mouse clicks, the status quo “works for me”. I’m all for paying the artist/author, I just don’t want the thing I’ve bought taken away from me as soon as the publisher decides to pull the plug on their DRM server.

    And what I’ve noticed here in Switzerland, even though it’s non-representative and anecdotal evidence, is that more and more ebooks are sold DRM-free.