• 0 Posts
  • 13 Comments
Joined 3 months ago
cake
Cake day: March 21st, 2026

help-circle
  • Name recognition counts for something in the choice about helper apps.

    Sometimes venerable old utilities you know from back in the day are bought out and turned into malicious shit. I don’t discount that.

    However when you go looking for a little helper app for something simple, there’s an ocean of weird little offerings out there and many of them are malware.

    I’d rather roll the dice that a venerable old classic hasn’t been bought out. This fact is probably quite Google-able. As for the long list of other unzip utilities… how am I supposed to know? Reviews and ratings are all fake. Many Reddit recommendations are fake.

    Just saying this is one angle on why people might continue using really old tried and true programs.





  • I agree with you about the tolerance paradox. I am sure you’d also agree that we must decide when to apply it because not every disagreement is a tolerance paradox.

    But now that we’ve named the tolerance paradox, now that it’s a meme we can just invoke on a dime, I find it gets used constantly. The word fascist is also everywhere now. Anything remotely draconian and you are a fascist and then a Nazi. The word Nazi gets thrown around very lightly around here. And the conversation slips very quickly from “don’t tolerate Nazis” to “it’s okay to punch them in the face” to “it’s a moral imperative to kill them.”

    I think we should be extremely judicious about deciding it’s a moral imperative to kill anyone. And I don’t see that extreme judiciousness around Lemmy generally. This political party’s beliefs are repugnant. And I am not ready to call for the death of someone who gives them money.

    I’m not saying I know exactly where to draw the line. But when I even try to engage people on where the line should be, I quickly find myself on the other side of it in their eyes. I’m an apologist, a collaborator, a fascist myself, and so on.

    This climate overall is deeply uncomfortable for me and even worse is the lack of self reflection about it.

    Maybe my problem is thinking that discussion actually matters and trying to fully engage the topic, when others are just here to talk big and pronounce death upon their perceived foes.









  • My employer has this figured out the right way: AI tools in the hands of competent employees can get a lot done. The engineers themselves agree and are long past their initial chilly reaction. They now have their workflows figured out and routinely report that they did something in minutes that would have taken them hours to do manually.

    And even under this rosier picture, token budgets are an issue. These guys have agents running agents running agents that use a whole ecosystem of internal MCP servers and skill plugins and on and on. They say they never shut their laptops because they have 6 or 7 things running at all times. The LLM usage is ridiculous.

    Once the LLMs cost what they actually cost, there will be a big time reckoning. First companies will crack down, then they will futz with on-premise models, then they will drive for SLMs and other slimmed down stuff. And they will pay the big guys 100x what they are now for 1/100th the usage.

    But hey I don’t know why I’m posting in substance about the real world when all we are here to do is say AI is bullshit and CEOs are dumb. Carry on I guess.