Now introducing "lafn" -- Lame Ass File Navigator.
https://redd.it/1jpz7fy
@r_linux

GNOME & KDE Plasma Wayland Sessions Outperforming Xfce + LXQt On Ubuntu 25.04 For Linux Gaming
https://www.phoronix.com/review/ubuntu-2504-x11-gaming

https://redd.it/1jq72ip
@r_linux

Here's the latest quarterly progress report for Graphite, the FOSS 2D graphics editor I've been building for 4 years
https://graphite.rs/blog/graphite-progress-report-q4-2024/?utm_source=reddit&utm_campaign=linux

https://redd.it/1jq7m8d
@r_linux

WattWise: Terminal-Based Power Monitoring Using Smart Plugs
https://www.naveen.ing/cli-for-smartplugs/

https://redd.it/1jqfu2f
@r_linux

Chris's Wiki :: The order of files in /etc/ssh/sshd_config.d/ matters (and may surprise you)
https://utcc.utoronto.ca/~cks/space/blog/sysadmin/OpenSSHConfigOrderMatters

https://redd.it/1jqigtn
@r_linux

EU OS | Community-led Proof-of-Concept for a free Operating System for the EU public sector
https://eu-os.gitlab.io/

https://redd.it/1jpyo0s
@r_linux

It won't be EOL on Windows 10 that drives the world to Linux, it'll be these tariffs.

Tariffs equal more expensive laptops, which equals people opting for older machines, and older machines work terribly on Windows 11, but on Linux they work wonderfully, so Linux it is. Makes you start to dream a bit, picture a renaissance of OS minimalism, DWM and i3 trending on TikTok. Influencers rocking Hyprland.

https://redd.it/1jqnzfr
@r_linux

KDE Plasma higher power usage than Windows at IDLE?

Hi, I'm running Bazzite KDE Plasma as my main OS right now, having dual boot with Windows 10. I'm measuring power draw out of the wall using GreenBlue GB202 and what I've noticed is that - having no apps running in backround - just dekstop right after logging in, on Bazzite the power draw shows around 110 W, while on Windows it is around 90 W. This also corresponds to what is HWInfo64 and btop showing about power usage - Bazzite \~\~28 W idle, Windows \~\~9 W idle.
I also tried Linux Mint 22.1 Cinnamon, and it also shows after booting just around 10 W on idle. This power usage continues, eventually dropping to \~\~20 W at idle

HOWEVER, what is interesting - after some heavy gaming on Bazzite (like 4h straight using GPU for 95%+, ex. CP2077), when I measure the power of the GPU - it is lower at around 7-10 W.

Now the question is: is KDE just more demanding from GPU than Windows/Cinnamon? Is this some kind of driver issue, or maybe new GPU needs to get ran hot for some time (I bought it 2 months ago at the beggining of february).

My specs:

CPU: Ryzen 5 7600,
RAM: CL30 6000 MHz 32 GB,

GPU: Sapphire Pulse 7800 XT,

NVME: 1x gen 3, 1x gen 4,

2x SATA SSD's, 1x HDD.

resolution 1920 x 1080, 75 Hz.

Thanks for any reply.

https://redd.it/1jqshp1
@r_linux

First time deleting Windows and using Linux as main system. Wish me luck guys, in 2024 you showed me a new world.
https://redd.it/1jquehf
@r_linux

help with command /../

Hello!

Im quite new to Linux, and read a file location with /../data/cent/things

Now my question is, .. is upp one, but how does /../ work?

https://redd.it/1jr5vqy
@r_linux

What distro should I use for dual boot? (considering Arch)

I only own a windows machine, and have so far in my pentesting journey strictly used Kali Linux, more specifically mostly I sit in WSL for Kali (windows subsystem for linux) and solely work with the terminal for my projects and exploits, which I enjoy a lot. I have also used VMs for ubuntu and have some experience with that.

However, I would still call myself a beginner, since I have A LOT to learn still.

I am considering in getting a dual boot to get a more authentic and native linux feel, to learn more and to also be able to do some exploits that a VM might not be able to where you need to access hardware etc.

I think that I want this dual boot environment to be more aimed towards cybersecurity and pentesting work, but I wouldn't mind if it can be used as a daily driver for programming/other software engineering projects etc. as well.

I will of course use my windows environment for my real personal use, gaming etc.
Is dual boot a good choice for me, and what distro would you recommend me to get?

I am quite interested in Arch/BlackArch. Since that would really force me to understand how OS works under the hood, which might make me a better pentester which can increase my chances to land jobs in the future. So I could see it as an investment, but I'm also unsure whether it's really a good idea and if Arch/Black arch would give me too much unnecessary headache and s too much hassle, and if it could mess up my windows machine if I do something wrong?

Or maybe it's not as difficult as I imagine as long as I do each step carefully and read the documentation?

Can arch/blackarch be used as a daily driver for both programming and security projects?

Would you maybe recommend another distro for dual boot - like ubuntu, kali, garuda or something else?

All tips are appreciated, and thanks!

https://redd.it/1jr9l3y
@r_linux

Audio Routing system?
https://redd.it/1jrdemm
@r_linux

APK on Chromebook using linux

Im trying to install APKs on Chromebook using Linux, I've been following a guide and I've tried to use the command: adb devices
However when it says list of devices attached it shows nothing.
And when trying adb install after it says install requires an argument

What's the fix?

https://redd.it/1jrfyb3
@r_linux

How do you use GNU stow? Entire .config folder (stow .), or individual packages (stow bash nvim tmux)?

First, if you've never heard of GNU stow, it allows you to keep your config files in a Git repo, do git clone git@github.com:myusername/dotfiles, then run cd dotfiles; stow . and all your config files in your home directory are now symlinks into the Git repo.

But there are two ways to use stow. One is to create a "unified" dotfiles repo, which contains the same structure as your home directory (a .config dir, and some individual files like .bashrc and so on). Then after checking out your dotfiles repo, you just run stow . and all your config files are in place.

The other way is to create a directory in your dotfiles repo for each individual config you might want to use (GNU stow calls these "packages") and then pass the names of each piece of software to stow, like stow bash nvim lazygit.

Some examples might be in order. Here's what a "unified" dotfiles repo might look like:

dotfiles-unified/
├── .bashaliases
├── .bashcompletion
│   └── alacritty.bash
├── .bashrc
└── .config
├── lazygit
│   └── config.yml
└── nvim
├── about.txt
├── .gitignore
├── init.lua
├── lazy-lock.json
├── lazyvim.json
├── LICENSE
├── lua
│   ├── config
│   │   ├── autocmds.lua
│   │   ├── keymaps.lua
│   │   ├── lazy.lua
│   │   └── options.lua
│   └── plugins
│   ├── example.lua
│   ├── lush.lua
│   └── nvim-notify.lua
├── .neoconf.json
├── README.md
└── stylua.toml

8 directories, 20 files

And here's what a "packages-based" repo might look like:

dotfiles-packages/
├── bash
│   ├── .bashaliases
│   ├── .bashcompletion
│   │   └── alacritty.bash
│   └── .bashrc
├── lazygit
│   └── .config
│   └── lazygit
│   └── config.yml
└── nvim
└── .config
└── nvim
├── about.txt
├── .gitignore
├── init.lua
├── lazy-lock.json
├── lazyvim.json
├── LICENSE
├── lua
│   ├── config
│   │   ├── autocmds.lua
│   │   ├── keymaps.lua
│   │   ├── lazy.lua
│   │   └── options.lua
│   └── plugins
│   ├── example.lua
│   ├── lush.lua
│   └── nvim-notify.lua
├── .neoconf.json
├── README.md
└── stylua.toml

12 directories, 20 files

The advantage of the "unified" approach is that you just have to run stow . and all your configs are in place. The disadvantage is that now ALL your configs are in place, including some configs that might be machine-specific (you might not have the same software on every machine, for example).

The advantage of the "packages-based" approach is that you can pick and choose: if on one machine you use fish while on the other one you use bash, you can run "stow fish" or "stow bash" and only the appropriate config will be put in place. The disadvantage is that it's more complicated: instead of running "stow ." and having all your configs in place, you have to run "stow package1 package2 package3" and you might forget one. (Or you have to create a per-machine shell noscript and put that in your dotfiles repo; either way, it's an extra step).

Those of you who use GNU stow, which approach did you choose? The unified "all configs at once" approach with stow .? Or the package-based approach where you have to run stow bash lazygit nvim but you can keep different machines' configs all together? Also, why did you choose the approach you chose, and why do you like that one better than the other

approach?

https://redd.it/1jrf8c4
@r_linux

How to launch a gtk binary in tty?

Right now I have an GTK binary (github) that I programmed to be a display manager and I need to launch it in tty as a service. Right now i took the sddm service file and made it launch my binary instead and that did not work. Any Ideas?

https://redd.it/1jrit4p
@r_linux

Void Linux just became the most based rolling release distro for me after reading this
https://redd.it/1jrkn3q
@r_linux

I Fought KDE Bugs for Weeks. My Cat Solved It in Seconds

I've been using Linux since 2017. My first and daily distro has always been Debian with Xfce. While I’ve mostly kept my setup pretty conservative, I occasionally get the urge to try something new.

A few weeks ago, I decided to experiment with the latest version of KDE. I started with KDE Neon, and while I really liked the look and feel, it turned out to be quite buggy. The screen would break in various ways, there was a lot of tearing while watching YouTube videos, Discover had weird blue lines, and there were many other visual issues. So, I decided to go back to Debian.

This time, for a bit of a change, I went with Debian Testing and installed the KDE Plasma version of Trixie. I was aiming for that “latest and greatest” experience on Linux, something I’ve generally avoided, because every time I’ve tried a rolling-release distro, it ended in disaster after an update. Debian Testing felt like a good compromise: newer packages, but not completely bleeding-edge.

At first, things were great. I could watch YouTube, write programs, and run local LLMs with performance similar to what I had with Xfce. But then the problems started creeping in. Microstuttering appeared, and games became nearly unplayable. Counter-Strike 2 wouldn’t even run properly, and Garry’s Mod turned into a brown visual mess.

YouTube began stuttering constantly, regardless of resolution or framerate. Some videos even played with audio while the video stayed completely frozen. I also started seeing interlacing artifacts in Kdenlive and VLC. To make it worse, text across the desktop began to look blurry and garish, like I was reading it through an old RF cable connection. (If you ever used 8-bit micros in the '80s, you know exactly what I mean.)

I was considering trying out a GNOME-based distro like Zorin OS, or maybe finally giving Fedora another shot. (I hadn’t touched it since Fedora 34, and back then my machine really struggled with it.)

So, I started downloading Fedora. While it was downloading, I went to make myself a coffee. Meanwhile, my screen locked... and my cat, being the helpful creature he is, decided to sit down on the keyboard.

When I came back and unlocked the machine, I noticed something strange: all my programs were closed, and, miraculously, the text on my screen looked perfectly crisp. I had previously tried everything in the book to fix that issue, and nothing had worked.

Confused but intrigued, I resumed the Fedora download and also grabbed OpenMandriva, just in case (I have another machine running it beautifully).

Then I started testing things:
- Kdenlive ran smoothly.
- YouTube had zero stutters.
- Steam actually opened at the right size! (Did I mention how, before, Steam looked comically tiny, and trying to scale it up somehow made it even smaller?)

Suddenly, everything was working perfectly.

I was scratching my head, trying to figure out which update could've fixed it... until I opened the system settings and checked About This System. Right there, it said:

> Graphics Platform: X11

It all made sense. Wayland had been the root of all my problems, and somehow, my cat had unknowingly switched the session manager to X11... and fixed everything.

EDIT: AMD CPU + AMD GPU. So I cannot just blame Nvidia on it

https://redd.it/1jrmisd
@r_linux

Linux 6.15's New "hugetlb_alloc_threads" Option Can Help Speed-Up Boot Times
https://www.phoronix.com/news/Linux-6.15-MM

https://redd.it/1jrmtok
@r_linux

feeling nostalgic

I am feeling rather nostalgic today and started reminiscing about the old school distro Mandriva One (from 2009). That was my first long term distro, longer than Mandrake and longer than RH, prior to migrating to Fedora 10, where I stayed until they upgraded the package manager from YUM to YUMI.

I was then on Simply Mepis for a while, but then I moved to Debian-based distros -- first Ubuntu, then a handful of other distros, such as Linux Mint, before finally settling on Parrot Security OS (circa version 4.7), and I am now writing this from Parrot Security OS version 6.3, which has become my favorite distro over the last 6 years.

Humor me -- what distros have you used that you look back on with fondness and miss using? Let's show some love for the older distros!

https://redd.it/1jrp2sr
@r_linux

Would it be feasible to create a new world wide web?

This is something that has been on my mind for a while now, apologies in advance if this is the wrong place for it.

Basically, even before AI the web had become flooded with extremely lazy and often misleading content. This exists solely for generating ad revenue, providing nothing of value and just making it harder and harder to find what you're actually looking for. Without ad engines like Google Ads, none of these sites, posts, videos, or whatever would have a reason to exist.

So my idea is to do something along the lines of creating a new internet protocol based on HTTPS, but with certain functionality explicitly disabled like cross-site content loading and third-party cookies. Along with that would be browsers and search engines that support only this new protocol. Ads would have to be a static part of content, and this would raise the barrier of entry to securing ad revenue, especially for low-quality content.

Obviously it is possible to add these restrictions to current browsers, but that does nothing to filter out all of the crap that's out there. So that's why it would need to be a completely new space. I suppose any existing content-hosting platform could add support for the new protocol, and carry with it all its existing junk. But there would be little incentive to do so.

I don't know, I'm just sick of it. AI seems to be accelerating the problem exponentially.

https://redd.it/1jrs1he
@r_linux