Post #4358827

It is becoming difficult and counterproductive, when I am seeking answers about Unix and its offshoots, to not simply read google's AI answer. Especially FreeBSD. It puts the command you need right there where you cannot ignore it, and at that point, it is ridiculous to not just try the command, as long as it's not piping curl to the shell or any such nastiness. It has not been wrong yet, and I have tentatively thrown some detailed Sysadmin tasks at it that would have taken me all day to figure out, stuff where nobody has solved and posted the specific problem&solution yet, and I'm still rusty with BSD in general, so it's one of those "it's time to sit down and RTFM and learn how to do the lower-level Unix thing," because there is documentation for that but no specific set of steps for you to mindlessly follow. I generally prefer to mindlessly follow a trusted source, I know that's bad but it's like Twinkies, my only hope is keeping it out of my field of view. Anyways, the fucker has ingested that documentation, and so, since FreeBSD is incredibly well-documented, and not subject to the rainfall of outdated or simply wrong tutorials that Linux provides for you to sort through, it is actually pretty good at writing sysrc files, for instance, if a given port or pkg doesn't provide them for whatever reason. What this tells me, is I need to figure out local inference asap, cause I am not going to keep using datacentre product, but, as I always expected, AI is very good at tech stuff, better than it's going to be at anything else. I guess it's also important to remember that I consider my skill level at this to place me in the "Centaur" category, when it comes to using AI assistance on tasks. I do have the discernment to spot a very bad thing to put in your shell, and also if something seems to be barking up the right tree. AI will not replace human skill, but it certainly can augment it and speed it up.