• 0 Posts
  • 17 Comments
Joined 1 year ago
cake
Cake day: June 8th, 2023

help-circle





  • Syslog (rsyslod) is usually the standard answer for the average sysadmin, but it depends a lot on your needs. A lot of newer loggers output as pure JSON, which offer benefits to readability and more approachable search logic/filters/queries (I’m so tired of regex).

    When you start venturing down the road of finding the right way to store and forward the output of logging drivers from Docker containers, as one example, rsyslod starts to feel dated.

    The easy answers if you want to throw money at the problem are solutions like Splunk, Datadog, or New Relic. If you don’t want to (and most people wouldn’t), then alternatives certainly exist, but some of them are just as heavy on system resources. Greylog has relative feature parity with Splunk Enterprise, but consumes just as much compute and storage if not more, and I found it to be a much larger pain in the butt to administer and keep running.

    The likeliest answer to this problem is Grafana Loki, just based on what I’ve read of its capabilities, but I haven’t had a chance to circle back and test it out. Someone here who has might be able to weigh in and speak to its strengths/weaknesses.



  • The logical fallacy here being that, based on that context alone, you should care because you will have something to hide in the future. Saying you have nothing to hide is always used in the context of one’s sense of guilt, or lack thereof, based on past actions. A counterargument would then be to ask why you should be allowed to hide your future wrongs.

    For many, the subject has nothing to do with that. It’s about not wanting to be monetized without consent. There’s also benefits in the form of protection against identity theft or social engineering. For others, the simple right to fundamental personal privacy itself is important - it’s about not having all of one’s life’s details on public display.

    Also known as “none of your goddamn business.”

    As a tangent, because it’s now stuck in my head and needs expression - the more thought you give to the problems introduced by technology that blur or step over this line, the more you realize how much harder it’s becoming to prevent outcomes where privacy is lost.

    Only engaging AI under tightly controlled circumstances is one thing; having it in the background perceiving everything you say and do on your desktop is a very different conversation. No matter what assurances are given that your privacy is protected, almost every situation like it that’s arisen since the advent of personal computers has resulted in a loss of control through duplicity, intrusion, sabotage, bad design, or floundering integrity.




  • This is the most insidious conundrum related to AI usage. At the end of the day, a LLM’s top priority is to ensure that your question is answered in a way that satisfies that model. The accuracy of its answers are a secondary concern. If forced to choose between making up BS so it can have a response that looks right versus admitting it doesn’t have enough information to answer, it can and often will choose the former. Thus the “hallucination” problem was born.

    The chance of getting your answer lightly sprinkled with made up stuff is disturbingly high. This transfers the cognitive load of the AI user from “what is the answer” to “I must repeatedly go verify everything in this answer because I can’t trust it”.

    Not an insurmountable obstacle, and they will likely solve it sooner rather than later, but AI right now is arguably the perfect extension of the modern internet - take absolutely everything you read with at least a grain of salt… and keep a pile of salt cubes close by.



  • sLLiK@lemmy.mltoArch Linux@lemmy.ml*Permanently Deleted*
    link
    fedilink
    English
    arrow-up
    1
    ·
    edit-2
    1 year ago

    Arch, i3, GTX 3080 12GB, and no issues. I’m holding off migrating to Wayland for the sake of full compatibility with all screen-sharing solutions.

    I’ve never really experienced any issues pairing Linux with nVidia, so I have trouble personally relating to all the hate they catch. There have been a few times where the kernel and the nVidia driver were mismatched, which caused issues trying to start up Xorg, but that’s easily solvable.


  • One of the main reasons my wife hasn’t taken the Linux plunge is Photoshop support and a lack of feature-complete alternatives with sane UI design choices. We would gladly pay for a Linux version of Photoshop at this point.

    It"s dawning on me now as I write this that Proton could be the secret sauce that slays this monster. Has anyone tried adding Photoshop as a non-Steam app to the Steam client, lately?