Daemon Silverstein

I’m just a spectre out of the nothingness, surviving inside a biological system.

  • 0 Posts
  • 16 Comments
Joined 1 month ago
cake
Cake day: August 17th, 2024

help-circle
  • According to my searches, while a RTG uses radioactive material weighting in the scale of kilograms (average of 5 Kg across missions such as Voyager and Cassini), a nuclear power plant requires several tonnes worth of plutonium and enriched uranium. The minimal critical mass for plutonium is 10kg, the double of how many fuel RTGs hold (that’s why RTGs don’t blow while ascending and/or on space). It’s a large difference of mass/weight between RTG fuel and rods for nuclear power plants. They’d need to carry the whole tonnes worth of radioactive material split across very small quantities (which would require a lot of lead walling and/or launches)




  • I’m a 10+ (cumulative) yr. experience dev. While I never used The GitHub Copilot specifically, I’ve been using LLMs (as well as AI image generators) on a daily basis, mostly for non-dev things, such as analyzing my human-written poetry in order to get insights for my own writing. And I already did the same for codes I wrote, asking for LLMs to “Analyze and comment” my code, for the sake of insights. There were moments when I asked it for code snippets, and almost every code snippet it generated was indeed working or just needing few fixes.

    They’ve been becoming good at this, but not enough to really replace my own coding and analysis. Instead, they’re becoming really better for poetry (maybe because their training data is mostly books and poetry works) and sentiment analysis. I use many LLMs simultaneously in order to compare them:

    • Free version of Google Gemini is becoming lazy (short answers, superficial analysis, problems with keeping context, drafts aren’t so diverse as they were before, among other problems)
    • free version of ChatGPT is a bit better (can keep contexts, can issue detailed answers) but not enough (it does hallucinate sometimes: good for surrealist poetry but bad for code and other technical matters when precision and coherence matters)
    • Claude is laughable hypersensitive and self-censoring to certain words independently of contexts (got a code or text that remotely mentions the word “explode” as in PHP’s explode function? “Sorry, can’t comment on texts alluding to dangerous practices such as involving explosives”, I mean, WHAT?!?!)
    • Bing Copilot got web searching, but it has a context limit of 5 messages, so, only usable for quick and short things.
    • Same about Bing Copilot goes for Perplexity
    • Mixtral is very hallucination-prone (i.e. does not properly cohere)
    • LLama has been the best of all (via DDG’s “AI Chat” feature), although it sometimes glitches (i.e. starts to output repeated strings ad æternum)

    As you see, I tried almost all of them. In summary, while it’s good to have such tools, they should never replace human intelligence… Or, at least, they shouldn’t…

    Problem is, dev companies generally focus on “efficiency” over “efficacy”, wishing the shortest deadlines while wishing some perfection. Very understandable demands, but humans are humans, not robots. We need our time to deliver, we need to cautiously walk through all the steps needed to finally deploy something (especially big things), or it’ll become XGH programming (Extreme Go Horse). And machines can’t do that so perfectly, yet. For now, LLM for development is XGH: really fast, but far from coherent about the big picture (be it a platform, a module, a website, etc).




  • Daemon Silverstein@thelemmy.clubtoLinux@lemmy.mlProblems with Arch upgrade
    link
    fedilink
    English
    arrow-up
    24
    arrow-down
    3
    ·
    17 days ago

    You didn’t specify which problem or which thing that broke. However (and based on my previous experiences on that matter), one could face a problem regarding package PGP/GPG signatures upon trying to update. This is because archlinux-keyring is not being updated before the signature checking. That said, a better approach is to always update archlinux-keyring (sudo pacman -S --needed archlinux-keyring) before anything else (sudo pacman -Syu). This way, you guarantee to be up-to-date with developer signatures, needed for pacman to check the validity for every package to be updated/installed. There’s also a pacman-key command, but I never had to use that.




  • If Linux is configured to use LUKS and/or Windows is configured to use Bitlocker, it’s not so simple as just installing the ext4/NTFS driver.

    Also, neither Linux can run Windows programs (I’m aware of Wine, but AFAIK Wine won’t run software already installed on an existing Windows installation) nor Windows can run Linux programs (I’m also aware of WSL, but apart from very specific chroot-ings, AFAIK one can’t run software from pre-existing Linux installations)…


  • Simultaneously, Microsoft has been expanding their efforts so to require Windows users to upgrade to Windows 11, even those who own old machines that don’t have TPM 2.0, while those machines are prohibited to really upgrade to Windows 11, meaning that their owners would need to buy another PC/laptop. Several Windows users were using a cheat to install Windows 11 without TPM 2.0, but Microsoft has been patching it, so it’s going to be a no more. Users of Windows 10 will have two options: buy another PC or migrate to Linux. I’d bet Microsoft already knows the latter possibility. Several distros generally come with the option “dual-boot installation” as default, so there are many novel Linux users, migrating from Windows, that chose to keep Windows together with Linux (so to not lose files and configs they made on Windows). What if something broke Linux and these users that are trying to escape Windows are now forced to use Windows?


  • Back in the days I used to use Windows, I did use Linux as a developer sometimes, yet I was sticking to a daily usage of Windows… Until Windows 10, when Windows started to be aggressive on how it won’t let me control my own machine (e.g. I couldn’t disable updates the way I wanted, I couldn’t run some softwares, I couldn’t this and I couldn’t that). Then I said “enough” and started using Linux on a daily basis, firstly Ubuntu, then I started to experiment on other Linux distros, until I finally landed on Arch Linux, as it’s highly customizable and let me have full control of my own machine, not being stuck to specific DEs (I know that distros like Ubuntu allow the user to uninstall the current DE, or install other simultaneous DEs, but Arch comes without any DE from scratch). I’ve been using Linux on a daily basis for almost a decade now and I don’t miss Windows.





  • There are officially 193 countries, according to UN. Each country with their own laws, some of them (European) with common laws (EU laws). How is it humanly possible for a site to keep track of every single law or every single country? Laws are not a worldwide consensus. Also, who and what exactly defines what “misinformation” is? For example: the belief in the supernatural (such as the daemonic forces from Göetia and Luciferianism) is not a scientifically proven thing, so, if we consider “non-misinformation” the information that is capable of being strictly proven, then should absolutely every social network content regarding one’s belief be considered “misinformation”?