• 0 Posts
  • 220 Comments
Joined 1 year ago
cake
Cake day: June 29th, 2025

help-circle




  • The thing about AI is, that this kind of concentrated control over a technology is more baked into it than with other internet things.

    You need control over heaps of training data to get anywhere with it and you need a lot of computing power to train it and to run it.

    Not impossible to do for smaller players, but a lot harder to compete with the giants.


    And on top of that it kinda clashes with traditional open source code sharing. The code alone isn’t helping you much, if you don’t have access to the above.

    Yes, you can copy and share a trained AI model, but that’s not the same as sharing code and it is fundamentally intransparent. So in a way it automatically violates some of the free software principles. Which might also rub the Lemmy crowd the wrong way, since many people here lean towards free software.




  • Well, yes. I’m not arguing against your general idea. Just the notion that laws would protect you.

    But strictly speaking, the internet isn’t a requirement for automated mass surveillance. Powerful enough computers to analyze and process our phone calls would have sufficed. The internet just happened to predate those.

    And yes, it’s always also a matter of ressources. But if you rely on that, your privacy depends on the assumption, that you’re just not interesting enough to be surveilled. Since you had no way to verify that, relying on that notion is somewhat similar to the “nothing to hide, nothing to fear” argument today and yields the protection of your privacy to the decisions of the surveillant.



  • Sort of an expert here: Could be decades, could be months, could be days.

    It all kinda depends on geological conditions (as well as the condition of your septic tank). Groundwater flow rates can be very different. If you have sandy soil, it will be a lot quicker than if you have a lot more silt or clay.

    It should be no less than 50 days though, because in that timeframe it is usually assumed, that microbial activity in the soil as well as adsorbtion can remove most of the harmful substances.

    I hope whoever built that system took that into account, though.







  • SpongyAneurysm@feddit.orgtoToday I Learned@lemmy.worldPoutine
    link
    fedilink
    English
    arrow-up
    1
    arrow-down
    1
    ·
    23 days ago

    Oh you, who art so wise in the ways of poutine, please enlighten me!

    I’ve heard praise of that stuff, but you can’t get it here in Germany, so I’d have to make it myself. What would be your advice/recipe for a decent DIY poutine?

    I suppose the hardest part will be finding an acceptable type of cheese. Afaik cheese curds aren’t even produced here. Do you know anything that’s similar enough to be passable? What even are cheese curds? It’s not that we lack variety in dairy products in general, but I’ve never encountered them, or maybe there’s something getting lost in translation. My dairy vocabulary might not be up to the task.


  • Wales though, isn’t going to be at the heart of the heat wave. Them having oceanic climate and being pretty far north already, they won’t be as hot. But I was surprised to see, that even there it’s going to be more than 30°C for a bit.

    Other countries like Spain, France and Italy are gonna experience the real extremes, with temperatures as high as 42°C in some places. Fortunately, AC is a bit more common there, as they are experience higher temperatures more frequently, but still.

    In Germany, where I live, we’re not very well prepared for the prolonged heat, and AC is still pretty rare here, especially in homes (and somehow has a bad rep).

    We have a culturally ingrained technique to deal with it, by blocking out direct sunlight with our window blinds (if you haven’t seen our common blinds, watch this: (https://www.youtube.com/watch?v=_KePLmie7lE) and airing out the house at night to cool down the building. This worked pretty well for the climate that used to be. But for heat waves like the one in front of us, where it doesn’t cool down below 25°C we’ll reach our limits.

    I’m so happy to live in an appartement, that recently had split heat-pumps installed to replace it’s old school electric heating system. Those can be used for cooling as well.


  • Chalk it up as a lesson learned. There are some hoops you have to jump through, when installing Linux on a machine built and set up to use windows. One of these is secure boot.

    IIRC the older surfaces have some challenges. But usually, there are just a few switches you have to set right in the BIOS/UEFI-interface and then some things to be aware of when installing Linux (that don’t really have anything to do with being a windows machine, but with older hardware). One thing is to be aware, i& your running it in BIOS or UEFI mode. Afair, that has implications from the partition table on your system drive down to the possible partitioning-schemes of some system folders.

    And I don’t have experience with surface laptops, but I don’t think you’d need any obscure kernels from github. The mainline Linux Kernel should be fine, if you set it up correctly. And you shouldn’t be using Kernels from somewhere else, if you don’t absolutely know what you’re doing. (And even then, you’d be better off compiling it yourself, with the right settings).

    Usually though, it’s just a matter of having the right kernel-params set during boot and loading appropriate models, which can all be done with a stock kernel, either by using the right boot-flags or even on a running system.

    Compiling kernels, to my understanding, is something for Gentoo users, tinkerers and embedded system devs. Or if you want a streamlined installation on your laptop with minimal overhead; though that already puts you in the tinkerer category in my books.



  • You talked about a DIAAS (digestable indispensable amino acid score!) - you are my favorite person today.

    Actually, I didn’t. I just realized, that I linked to PDCAAS, which is a slightly different method. But it didn’t really matter, as I just wanted to illustrate the concept. And I’m not too involved in the topic. I don’t know what you’re doing, that you’re whipping up tables about this stuff, but I’m just a layperson with a little knowledge about nutrition.

    Allow me a few remarks though:

    1. you’re referencing prepared meat, but raw tofu? In my experience, tofu is usually also prepared in some way, and with most preparations, it looses quite a bit of water content.

    2. We don’t have to rely on mostly unprocessed plant food. There’s stuff like texturized vegetable protein, that delievers a more concentrated source.

    3. While a table like this gives a good overview and reference, it’s easy to miss the fact, that we usually don’t get our protein from a single source. As I mentioned in my last comment, combining different sources can be a good way to enhance the overall protein quality.
      To reflect that, we’d need DIAAS data for prepared dishes, meal plans or a whole diet.

    4. Can you even use DIAAS to calculate an amount of single protein source food, that you’d have to eat like that? I don’t know if it scales that way, and even if it did, for an incomplete protein source, you’d end up with a lot of excess for the abundant amino acids in that protein source, which I suspect would have to be excreted and I don’t know how your kidneys like that.

    5. That table would really benefit from adding references to clarify what you base your assumptions on and where you get your data from.

    But I think none of this is all that relevant for the underlying topic in this thread.