I’m for being against LLM, but I am also for using logically sound arguments. I find a far better argument against using it as a developer is: https://arxiv.org/pdf/2506.08872 (tldr: thinking is good actually!)

Some logical fallacies I see in that writeup:

• Gerrymandered definition
• False dichotomy
• Strawman
• Slippery slope
• Appeal to authority

I won’t go into all of thwm cause most are boring and I’m typing this on my phone. But here is some of my thinking.

This one irks me personally the most:

“As a lifelong tool builder and user, I know what tools are.” This is an irrelevant conclusion. Expertise in building tools doesn’t settle a definitional question, and the essays argument still has to stand on its own criteria.

For now, lets accept his definition of a tool and some of the arguments he uses and apply them to some other aspects of computing:

A computer is not held in the hand in any meaningful sense, transmits none of your force, and involves no continuous direction. You enter an expression and it “acts on its own.” By Hickey’s definition, a comouter is “automation for avoiding arithmetic.” or “automation for decoding a binary stream from the internet wire and rendering it in your own mind as a website”. Nobody that uses computers has hesitated to call it a tool for computation. And nobody thinks it makes the user less of a professional while using them.

Lets do another one, closer to the topic of programming: A compiler. You hand it a human readable detailed description of a program (source code), it acts entirely on its own, and under Hickey’s logic it is “automation for avoiding translating to machine code.” Yet the compiler is the pardigmatic programmer’s tool, and nobody claims programmers who use compilers failed to learn programming. If “acts on its own after a request” disqualifies an LLM, it disqualifies half the toolchain that built the software the essay was presumably written in.

And as a bonus, programmers usually don’t even read the output of the compiler to verify it’s correctness! See: Reflections on Trusting Trust by Ken Thompson https://dl.acm.org/doi/10.1145/358198.358210

Reading the title I thought you want to set up an analogue radio station and got all excited.

Clasic goomba funnel fallacy

The only thing I cared about was RAM amount cause I have some beefier things I wanted to run (multiple game servers that need 8-16gb ram each etc.) and that was the most affordable option.

Without storage, maybe thats the reason for the cheaper price, but luckily I had some 1tb NVMEs hoarded from the before times.

It has some linux problems, like I couldn’t make the NPU be usable, even though I used the ubuntu image that is supposed to support it, but those issues don’t matter for my usecase

I don’t think it’s availabke anymore: minisforum ms-r1

Fair point.

My thinking was: if OPs screw is best screwed in with a philips screwdriver, I won’t suggest a flathead one, just cause we’re in the flathead sublemmy. But thinking about it now, I got my priorities mixed up here, privacy goes over convenience for the audience here including myself.

Here are some selfhosted solutions that would function similarely:

with api:

• https://github.com/gristlabs/grist-core
• https://github.com/audreyt/ethercalc

without api:

• https://github.com/CollaboraOnline/online
• https://github.com/ONLYOFFICE/DocumentServer
• (probably eurooffice now as well right?)

I honestly believe a well made spreadsheet would be best.

• You can have it in “the cloud” for multiuser.
• You can put some automation around it through an API if you use google sheets or onlyoffice or one of the many other spreadsheet providers that have an API. (Like for example phone notification when you need to water or fertilize a plant)
• It’s easy to get started.
• And it’s infinitely flexible to suit your exact needs.
• You can probably even abuse a page and make a graphical layout map of your garden with clickable links to specific plant pages/sections.

I got myself a whole new ARM mini pc for 600$ with soldered 64GB of ddr5 RAM. I think of it like a very beefy rpi5 and use it as my homelab server. They were not very popular and selling leftover stock for the same pre-bubble price.

The amount of patches is as bad of a metric for judging development pace or quality of the result as counting the lines of code.

Vrei să pleci dar nura nura iei

AI-powered security platform

mh…

removing the commercial barrier can dramatically accelerate the adoption

*starts projectile vomiting at this startup culture techbro bs

It’s a subculture phenomenon, and lemmy linux is a subculture of a subculture, so it’s more intense.

I’ve been a moderator on some linux reddits in the past. And the less users there were in the sub, the more negativity, rudeness, gatekeeping and other toxicity was there. But ofc what I say is just an opinion based on my subjective experience so idk.

The linux kernel bsod is even younger than that, august 2024 iirc.

Hey, I happened to have a handful of old smartphones that still have an ok battery and just collect dust in the shelf.

If you want me to ship one to you, pm your address.

Gonna help my linux buddies so that they can scan QR codes if their OS crashes.

(not sarcasm)

with dokplpy as a server docker managementweb gui.

for the code push workflow:

• keep your normal git workflow, commit & push
• set up a git hook in git & dokploy that rebuilds from source on every push
• or do it on a schedule and build once a day or whatever

If you’re project has an unusual way to build, you could also do a git action that pushes a docker image out somewhere, depending on yoyr git provider.

There are probably infinitely many ways to autodeploy something.

If you need a step bt step breakdown, I can give you a guide based on my personal website, but that might not apply to your project.

On lemmy you’ll allways find the weirdos and elitists that have been pushed out of the mainstream.

The linux equivalent of bsod has been introduced pretty recently, so it’s normal for few people to have experienced it yet.

Also somewhat relevant xkcd: https://xkcd.com/1053/

I’m a gui pleb, so I do all my selfhosting through dokploy, which uses traefik under the hood and I’m quite happy with it so far.

from the top of my head there is a file ~/.config/dontaskagainrc that you can manually edit. I’m sure it’s somewhere in the settings gui as well but no idea

https://github.com/fluxerapp/fluxer

not sure about the security tbh. but it’s very discord like

Funny thing is, I have 2 keyboards and a standalone numpad and one mouse, but one of the keyboards registers itself in the system as keyboard and mouse, because it has mouse moving hotkeys.

So my system literally thinks I have 3 keyboards and 2 mice.

• I use a split ergo keyboard for typing. (this one has the mouse stuff)
• A 65% keyboard for gaming (so I have lots of room for the mouse and don’t have to change the keybindings for the split one)
• A numpad off to the left so it doesn’t block mouse space, need it for some games and spreadsheets.

After a work day is over I shove the split keyboard under the monitor and pull out the gaming one and place the mouse differently.

People that visit always comment on how weird that is, but it works for me.

https://www.borgbackup.org/

Not only does borg have strong encryption, it also has git-like incrimental versioning and insanely good compression and de-duplication.

For example, I backup my whole 700GB OS drive (personal data + the whole linux OS install with all my tweaks and configs) - I set it to keep the last 15 “snapshots” (7 daily, 4 weekly, 3 monthly, 1 yearly) and all it takes on the borg server is 800GB.

I like using borgmatic for automatic borg backups, but it’s optional.