• 0 Posts
  • 38 Comments
Joined 3 years ago
cake
Cake day: July 10th, 2023

help-circle
  • Sure, my argument is that not the modality of the sense matters, but the circumstances of that power. The resolution, the filtering and the agency someone has over that experience.

    Smell compared to sight would also cause generally less of a brain overload, since smell is low bandwidth compared to sight. So if future sight as a curse might be worse than future smelling.

    Doubling the amount of information you sense just in two moments, now and the most likely future in 3 minutes, might be more bearable with smell.


  • All the same situation would happen with future sight. My point is just that if you assume that future sight has resolved those questions, situation and paradoxes in a beneficial way, because it is a super power, the future smell would be a super power as well, and not a curse…

    If future sight has resolved them is a bad way, then it would be a curse, just as any other future sense.

    Just imagine always having to see all possible sights three minutes into the future at the same time. There always is a miniscule chance of all air molecules bunching up in the corner of your room or you falling through the floor…




  • How is smelling smells three minutes into the future a curse?

    I mean if it works similar to other time powers like seeing into the future, that future would be preventable.

    Smelling the future and acting to it seems like pretty awesome, and three minutes is pretty long… You have enough time to act and escape you death maybe, if you suddenly don’t smell anything anymore… Or you can make sure that you only eat the best food available… Since tasting food is mostly about smells… You could become an awesome cook… you just have to deal with the five tongue senses…


  • In case of illegal numbers, intention matters. Because any number could be converted to different numbers, for instance through ‘xor encryption’ different ‘encoding’ or other mathematical operations, which would equally be illegal if used with the intention to copy copyright protected material.

    This was the case previously. You cannot simply reencode a video, a big number on your disk, with a different codec into another number in order to circumvent copyright.

    However, if big business now argues that copyright protected work encoded in neuronal network models is not violating copyright and generated work has no protection, then this previous rule isn’t true anymore. And we can strip copyright from everything using that ‘hack’.


  • I had a similar thought. If LLMs and image models do not violate copyright, they could be used to copyright-wash everything.

    Just train a model on source code of the company you work for or the copyright protected material you have access to, release that model publicly and then let a friend use it to reproduce the secret, copyright protected work.



  • But you don’t need to misuse language to assign responsibility.

    What? I am interested… How else would you assign the responsibility to people that designed something intentionally bad, if you cannot used language?

    “Misuse [of] language” is a concept I cannot even begin to wrap my head around…

    Do I loose the warranty if I use language in unintended ways?

    It is their responsibility for breaking the system.

    You just ‘misused’ language to assign responsibility to people for breaking the system.

    Saying the system was always designed for this removes responsibility.

    No? Responsibility is not a binary concept. Someone can kill someone else, and would be responsible for that death, and the people around that killer could also share responsibility for not noticeing their unusual behavior. And the system could also be responsible for not giving the killer the support they needed, which drove them to kill someone. And the people that designed or constructed that system could also be responsible for not caring enough about these kinds of deaths to prevent them systemically.


  • There is a difference is saying “I does what it does” and “what it does is per design”. The latter assigns a responsibility.

    In OP Aziraphale gives socienty the responsibility to fix a broken system incrementally and Crowley gives the people in power the fault of intentionally creating a bad system and calls for revolution.


  • Hmm… I am using git for maybe 15 years… Maybe I’m just too familiar with it… and have forgotten my initial struggles… To me using git comes natural… And I normally pay a lot of attention to every single commit, since I started working on patches for the Linux kernel. I often rebase and reorder commits many times, before pushing/merging them into a branch where continuity matters.



  • Isn’t it the exact opposite?

    I learned that you can never make a mistake if you aren’t using git, or any other way for having access to old versions.

    With git it is really easy to get back to an old version, or bisect commits to figure out what exact change was the mistake.

    The only way I understand this joke is more about not wanting to be caught making a mistake, because that is pretty easy. In other methods figuring out who did the mistake might be impossible.






  • I know this post is more about the committing on LLM “fixes”, but find the other reasons more interesting.

    Similar to the date & time library there are a couple of other things that look easy at a first glance, but get complicated very quickly, because it has so many special cases:

    • lexicographic sorting (different languages sort things differently)
    • Postal address formatting (different standards in different countries, with many different context sensitive rules)
    • string handling