• 1 Post
  • 96 Comments
Joined 11 months ago
cake
Cake day: April 30th, 2024

help-circle









  • I don’t know what I’m reading.

    Here people mostly use debit cards, and we don’t worry about things like credit scores or whatever. Also these cards tend to be free. At the end, they just give you a convenient way of paying with money you actually own.

    Why would anyone buy normal stuff using credit? I get it for big purchases as a house or a car. But why go into debt for groceries or a new laptop?


  • The thing is that I already have a server and a few Terabytes of unused storage. So that would not be an issue. As long as storage doesn’t en up adding that much. I know that the fediverse protocol likes to replicate storage among all servers involved in an interaction. Though I wonder if it would be possible to safely erase old data, specially if I’m just hosting it for myself. I need to investigate on that.

    But for the other costs I already have a server running 24/7 on my house and several Tb of Storage. I already pay for that regardless as I use it for other things. Though ideally I would not want to allocate more than 500Gb for a one person instance, idk how reasonable would that be.

    And I also need to investigate how are the normal federation politics with one person instances. If it is like trying to host an email server would be hell as you’ll get mark as spam by a lot of providers.

    And now that I’m wondering things I wonder how feasible would it be to host very small instances on cheap devices like sbc or cheap mini-pc. Maybe aiming for thousands of instances with a few dozen people in each instead of a few dozen of instances with thousand of people in them.



  • Things doesn’t usually work like that.

    Rent subsidiaries work by your annual income and usually the cost of your rent.

    For instance they may pay you 300€ a month for your rent as long as your income is less than 30.000€ a year and your rent is bellow 800€/month. And increasing the thresholds if you have kids or if you are part of a protected collective.

    They may be above these thresholds. It’s pretty common in Europe for people who struggle to meet ends are above the needed thresholds for getting help. As prices have gone really up and the bar for being lower class have change a lot lately as there is a lot of new extremely poor people to help. So money don’t end up being enough for all. And people with normal jobs and who live alone or with their SO usually do not get any help even if they need it.

    The welfare state is kind of falling apart in the latest years.



  • Not every program is written for spacecraft, and does not net the critique level of safety and efficiency as the code for the Apollo program.

    I don’t even know. If memory issues are your issue then using any program with safe memory embedded into it is the way to go. As most things are actually made right now. Unless you are working in legacy applications most programmers would never actually run into that many memory issues nowadays. Not that most programmers would even properly understand memory. Do you think the typical JavaScript bootcamp rookie can even differentiate when something is stored in the stack or the heap?

    You are talking like every human made code have Linux Kernel levels of quality, and that’s not the case, not by far.

    And it doesn’t need to. Not all computer programs are critically important, people be coding in lua for pico-8 in a gamejam, what’s the issue for them to use AI tools for assistance?

    And AI have not existed before a couple of years and our critically important programs are everywhere. Written by smart humans who are making mistakes all the time. I still do not see the anti-AI point here.

    Also programming is not concrete, and AI is not sugar. If you use AI to create a fast tree structure and it works fine, it’s not going to poison anything. It’s probably be just the same function that the programmer would have written, just faster.

    Also, not addressing the fact thar if AI is bad because it’s just copying, then it’s the same as the most common programming texhnique, copying code from Stack Overflow.

    I have a genuine question, how many programmers do you think that code in the way you just described?





  • You can actually apply those tools and procedures to automatically generated code, exactly the same as in any other piece of code. I don’t see the impediment here…

    You must be able to understand that searching by name is not the same as searching by definition, nothing more to add here…

    Why would you care of the shit code submitted to you is bad because it was generated with AI, because it was copied from SO, or if it’s brand new shit code written by someone. If it’s bad is bad. And bad code have existed since forever. Once again, I don’t see the impact of AI here. If someone is unable to find that a particular generated piece of code have issues, I don’t see how magically is going to be able to see the issue in copypasted code or in code written by themselves. If they don’t notice they don’t, no matter the source.

    I will go back to the Turing test. If you don’t even know if the bad code was generated, copied or just written by hand, how are you even able to tell that AI is the issue?


  • Any human written code can and will introduce UB.

    Also I don’t see how you will take more that 5 second to verify that a given function does not exist. It has happen to me, llm suggesting unexisting function. And searching by function name in the docs is instantaneous.

    I you don’t want to use it don’t. I have been more than a year doing so and I haven’t run into any of those catastrophic issues. It’s just a tool like many others I use for coding. Not even the most important, for instance I think LSP was a greater improvement on my coding efficiency.

    It’s like using neovim. Some people would post me a list of all the things that can go bad for making a Frankenstein IDE in a ancient text editor. But if it works for me, it works for me.