• 1 Post
  • 185 Comments
Joined 3 years ago
Cake day: July 24th, 2023
  • fuck_u_spez_in_particular@lemmy.worldtoMicroblog Memes@lemmy.world•Prediction vs RealityEnglish
    1·
    1 month ago

    I know we’re fucked. But I refuse to believe that we cannot change. It will still get significantly worse, that I’m fairly sure of. But I also think that at some point for one renewables lobby is bigger than fossil and will actively fight fossilized lobby and that humans will see on a broader scale that we need to change. All of this is already in progress. I doubt doomerism does us any good. I’d agree that global collaboration would help significantly (and to a part this is happening already). Also keep in mind we’re prone to talk about extremes, like capitalism, communism, socialism etc. While a mix of those is more accurate for most states (social market economy). USA is probably much closer to capitalism, but I doubt their system will work much longer it’s too polarized to be “stable” much longer.

    So while it will get ugly (and let’s hope that climate scientists weren’t too optimistic) I think we at least may be able too avoid a runway hothouse earth and stabilize the climate to at least human survivability (which is still sad…)

  • fuck_u_spez_in_particular@lemmy.worldtoProgramming@programming.dev•LLMs are eroding my software engineering career and I don't know what to do
    8·
    3 months ago

    But I’m just another off-the-shelf engineer now. I have no domain expertise that another Sr. engineer steering an LLM cannot match.

    Funny because I think this is more important than ever. You can produce (ugly) code at remarkable pace now, being able to detect where it goes wrong is a very important skill nowadays.

    I still can’t let it loose too much because of all the garbage it still produces when given too much freedom. I’m honestly annoyed that my colleagues use it that much on production code, because it’s so much meh code with hidden (on the surface) inefficiencies and a lot of avoidable boilerplate and duplication.

    Then on the other hand I enjoy that it can create scripts (that are tedious to write) and prototypes at impressive speed. Generally prototyping and research has gotten quite enjoyable because you can stay most of the time on the theoretical side of things, quickly try things out or quickly get info to stuff you don’t yet understand enough.

  • fuck_u_spez_in_particular@lemmy.worldtoTechnology@lemmy.world•'People will buy intelligence from us on a meter': ChatGPT's Sam Altman's AI vision worries criticsEnglish
    2·
    3 months ago

    Right now it just makes more sense to not use the API directly, but the subscriptions, they seem to be better priced. Also, Anthropic seems to be more expensive…

    How that will look in the future, I don’t know, I’m fairly sure they’re progressively increasing the price, or rather reduce the amount of tokens you can use in the subscription (as they did with gpt 5.5 and claude opus 4.7). The chinese competitors are getting increasingly more interesting. It’s also quite impressive how well small models like qwen 3.6 27B run already on a (not so affordable) 24+GB GPU, unfortunately still far from the quality of say gpt 5.5, but probably comparable to o4 or something like that, certainly usable.

  • fuck_u_spez_in_particular@lemmy.worldtolinuxmemes@lemmy.world•Unironically at this point
    2·
    4 months ago

    So you haven’t had “zero” problems. AFAIK wayland is already usable with AMD since almost a decade or so… (well not every program was supported yet a decade ago obviously, but, at least these kind of issues that Nvidia has/had are non-existing AFAIK).

    It doesn’t fix the drivers but for many the installation and set up is where things go wrong. That’s how it was for me.

    For me it was never the installation, just the risk after updating the drivers that yet another issue appears (sometimes old ones were fixed though, to be fair).

  • fuck_u_spez_in_particular@lemmy.worldtoMicroblog Memes@lemmy.world•It's OnEnglish
    2·
    4 months ago

    I mean just stating the endless facts, would be enough for a somewhat sane person to grasp how fucking awful this current administration is. There’s endless fuel every day. Just collect a list of the worst stuff, you can still fill the whole day with a fairly neutral list of facts… That should be enough motivation to vote them away…

  • fuck_u_spez_in_particular@lemmy.worldtoProgrammer Humor@programming.dev•Now listen here you little shit
    1·
    5 months ago

    You will always be better at decisions than an n-dimensional matrix of numbers on an overpriced GPU.

    I’d be careful about these claims. Maybe with our current iteration of “attention-based” LLMs, yes. But keep in mind that our way of processing information is strongly limited compared to how much data is fed to these LLMs while training, so they in theory have a lot more foundation to be able to reason about new problems.

    We’re vastly more capable at the moment at interpreting our limited view on foreign code, being actually creative, find new ways to reason, yes. Capable developers (open source…) often have seen quite a bit more code than the average developer and are highly skilled, still with just a tiny subset of the code that an LLM has seen.

    But say these models improve in creativity and “higher-level of thought” through whatever means (e.g. through more reinforcement learning). Well, let’s just say I’m careful with these claims. These LLMs are already quite a help with stupid boilerplaty code (less so with novel stuff, and writing idiomatic non-redundant code, but compared to 2-3 years ago it’s quite a step already, to the point that they’re actually helpful, disregarding all the hype and obvious marketing strategies of these AI-companies)