• 2 Posts
  • 9 Comments
Joined 3 years ago
cake
Cake day: June 18th, 2023

help-circle




  • For me, the best proof that magic does indeed work in a sense, is the Pulitzer prize photo of the burning monk from Tibet, and in general the self-immolating budhists.

    It’s a proof that once you get your subconsious under reins, you can affect your life to an extreme degree. Budhists learn to do it through sheer training and willpower, western magic does it through rituals, belief and symbols, but I think the goal (and the mechanism that makes it work) is the same.

    Of course, it means that it’s limited just to things you can reasonably affect, so no i.e cursing people.

    But, if a ritual nudges your subconscious to i.e. make you study more instead of procrastrinating and you thus pass the test, it did work after all. It’s just not as flashy as people expect.

    This is my theory about how it works, and at least for me it makes sense. Plus, it’s fun!



  • You can do a task pretty well if you nudge the AI, have it write an exact explanation about every part of the architecture, code and data flow it’s working with and throw relevant files into context, and correct anything that’s wrong before you send it to do the task. You still have to review, but I didn’t have to correct much in my experience.

    But that burns like 20$ of tokens per task, at current prices that are way below the costs AI companies are paying.

    While it does help me, especially with parts of the codebase I’m not familliar with, it’s not sustainable, and it’s actively and very quickly robbing me of my skills and knowledge. It’s really a bad idea to use it, in two years time you’ll be royally fucked once they raise prices to recover the trillions they are loosing right now.

    So, however tempting, I simply don’t use it. I won’t throw away years of college and experience just to do a task a little bit faster today.




  • I don’t expect it to be balanced. I also don’t mind that a lot of users are centralized at lemmy.world - because we know the admins, and there isn’t a largescale corporation behind them that would be capable of monetizing and manipulating their userbase. The only kind of balance I want is that there isn’t an instance with multi-billion funding and teams of engineers dedicated to squeezing their userbase and monetizing every character they type. Sure, it will inevitably happen that someone from an instance would try something like this, but that’s an acceptable risk with community-run stuff and may or may not happen - and it probably won’t be as succesfull or large-scale. However, if you let in someone you are certain will do something like that, and has unimaginable amount of resources and manpower to do it, then they will just squeeze it dry while not shying away from exploiting every single privacy mistake there is in the protocol or our implementation.


  • I would like to also add this argument into the discussion, since I’ve seen a lot of people who are voting for federating with meta, with the argument that defederating just because we don’t like someone goes against the idea of Fediverse, and interconnected network of diverse servers that is should inclusive and allows people to connect.

    It’s quite the contrary - allowing Meta in goes directly against the idea of Fediverse, and we should fight it as much as possible.

    This is a literal quote from the main header on https://www.fediverse.to/

    The fediverse is a collection of community-owned, ad-free, decentralised, and privacy-centric social networks.

    Each fediverse instance is managed by a human admin. You can find fediverse instances dedicated to art, music, technology, culture, or politics.

    Join the growing community and experience the web as it was meant to be.

    Judging by this main selling point of the Fediverse, it sounds to me like Meta shouldn’t be in the Fediverse do begin with, and every instance should defederate from them by default.