From what I can tell xbps-src are just the source packages to the main repos in Void. That is not what AUR is. We have access to the main repo sources in Arch just like Void. The main thing about AUR is anyone can contribute without any gated approvals. That is the big difference between the main source repos of either distro and AUR. Unless I have misunderstood what xbps is.

but looking at templates they can actually understand its kinda simple script and get the idea of how it works

Same exact idea with PKGBUILDs. No benefit to Void here. The way Void does things will not change people looking at or understanding the packages they install. You have the same optitunities on both systems for looking at the source of packages. So that argument for Void is void :)

Also void has runit so this mean u have to get more simple programs to run system like seatd dbus and etc.

Not really a good argument either. Systemd and runit are different but that doesn’t make runit better in terms of learning anything. If you want to learn how most Linux systems boot and operate you need to learn systemd as that is what the vast majority of distros use. Learning runit instead only means you are learning a niche way of booting a tiny fraction of systems.

Neither of these arguments are a very strong case for Void over arch.

XBPS-SRC does not look like an alternative to AUR at all. It looks like Voids alternative to https://gitlab.archlinux.org/archlinux/packaging/packages - where Arch maintains all its packages. Nor is comparing the number of packages in AUR to Void main repos a good idea - Arch has its own main repos that are a better equivalent. The Void templates do not look dissimilar form what a PKGBUILD file is either and you can do the same things with writing your own PKGBUILD or pulling them from repos if you really want to. I don’t see how void is any better then Arch in anything you have described here. IMO it just looks like it does more of the same things with a bit difference in syntax/commands you run. Nothing you have said here is really a solid argument for using Void or Arch at all.

The AUR is not even that great. I think most people seem to get confused between what is in the AUR and the main packages since they just use tools like yay that install from both. But most people only use a couple of packages from the AUR - it is the package selection in the main repos which is what is so nice about Arch. The AUR is just nice for more niche things that have not made it into the main repos yet.

I hope u don’t use AUR blindly and just do yay -S something without looking what pkgbuild is doing, it might be dangerous not knowing what program can do and what script that is downloading it too right?

Same goes for Void? Most people wont read the source of third party packages they install. No matter what distro they are on. AUR tooling does try to help with this but most people ignore it. Same will go for Void. It is not a distro problem - just a humans are lazy problem. Plus even if people did read them there is only a small subset of people that actually understand them enough to spot obviously malicious packages - though that can spot hidden malicious packages are vastly smaller.

252 of that 592 used memory is buffers/cache, not application memory. That is used by the kernel for kernel buffers and the filesystem cache - IE files read by something at some point. The kernel keeps them in memory in case they are needed again to speed up file reads. You can effectively ignore these vales as they will always grow to fill your ram and will be evicted when programs require memory and there is not enough free.

These tools are not lieing to you, just telling you something other then what you are reading into them. Tracking and reporting on what is using memory is a complex topic and here used is just what is physically allocate. It doesn’t mean much over all as it always tends to be full of your system has been running for a decent amount of time. Available is typically the more useful one to look at as it is an estimate about how much the kernel can reclaim now if an application request it without needing to swap things out.

Can you share the output of free? There are multiple values to read from that.

None of that describes why a beginner would not want to use a immutable distro. It only describes why it is not a good idea if you want to learn how traditional linux distros work. Not all beginners want to learn or care about how linux works under the hood and immutable distros can be a good fit for them. They might not be right for OP, but you have not describe any reason why any beginner should avoid them.

I saw some people using Arch to learn the inside out of Linux, but I’m afraid It could be to challenging.

It will be challenging but IMO give it a shot if you think it is something you might want to do. No harm in trying really. If you mess it up or find it too hard or whatever you can always install something else afterwards. It is not like you are stuck with your first choice forever. The only thing you will lose is a bit of time and will gain a better understanding of things even if you cannot make it fully work.

I don’t agree that arch is not a beginner distro - it is a DIY distro that requires a lot of reading and willingness to learn and understand things. The arch wiki is an excellent resource for anyone (on any distro IMO) and well worth reading. If you are OK with that work then it makes a fine distro for anyone, beginner or not. It is not a distro for many people - again does not matter if they are a beginner or not. It is for people with a particular mindset. One that you might change over time or as you grow and learn more overall.

No harm either if you decide it is not for you. Play around with a few distros and try to find which one works best for you. There is really no one best distro. Just a lot of different things that appeal to different people and the only real way to find out which you like is to try them out.

But why won’t anyone think of the AI shareholders…

If you want to be safe backup anything you care about before hand. Assume that you might end up needing to wipe the system and reinstall everything. Encrypting the Linux partitions should not affect windows but there are so many other steps that could go wrong when partitioning and installing a system it is better to backup everything you care about just incase.

--asdeps also works when installing something to immediately mark it as a dep. Can be useful for non dep packages if you only need it temporarily as it will be removed the next time you purge unused deps.

Clean orphaned dependencies: sudo pacman -Rs $(pacman -Qtdq)

In addition to this, or rather before, you can run pacman -D --asdeps package_name to mark a package as a dep. If it is no longer required by something else it will be removed with the above. This can be useful for things that are deps that you installed manually at some point for some reason.

And remember that you can recover from anything, even removing base packages or bootloader ones with a live cd and chroot or using pacman with a different root with the --root /mnt flag to pacman.

Otherwise if your system still boots it is just a matter of following the install instructions for whatever is not working like you did the first time.

I would not worry about virtual memory usage. Virtual memory can include memory mapped files and does not indicate actual ram usage - only the address space that the program has opened at some point. There is little point in worrying about it.

IMO the best thing is to just start using it. You will start to pick things up fairly quickly then. Puzzles don’t often ingrain different ways todo things and often focus on weird or niche things that don’t come up as often. They can be a nice supplement to not a substitute for just using it in real world usescases.

I do also find it helpful to read the shortcut keys on their site to get a feel for what is available. You won’t remember everything but it can be useful to know what is possible. Then when you hit a problem you may remember reading about something that can help and go look it up again.

Just dont format the drive when installing a new distro. BTRFS or not you can delete the system folders manually first if needed but I believe that some if not all distros will delete the system folders for you (at least ubuntu used to do this last I tried). And if not you can do it manually.

It does not matter if you have a separate partition or not for /home installers won’t touch it if it already exists except to create a new user if needed. Remember, all the installers do is optionally format the drives, mount them then install files into those drives. If you skip the formatting and manually do that partitioning (or using an existing partition layout) it will still mount and write to the same places regardless of it they are separate partitions or not. So a separate partition does not add any extra protection to your home files at all.

But regardless of what you do you should ALWAYS backup your home data anyway. Even with separate partitions or subvolumes the installer can touch or delete anything it wants to and you can easily click the wrong button or accidentally wipe thing. At most preserving your home saves you from restoring from a backup it should not be done instead of backup.

There is no problem with having home on a different disk. But why do you want swap on the slower disk? These would benefit from being on the faster disks. Same with all the system binaries.

Personally I would put as much as possible on the faster disk and mount the slower somewhere that the speed matters less. Like for photos/videos in your home dir.

/boot can be anywhere though if you are getting a grub error that suggests the UEFI firmware is finding grubs first stage but grub is having issues after that. Personally I don’t use grub anymore, systemd-boot is far simpler as it does not need to deal with legacy MBR booting.

What you have listed is a good start. https://www.rust-lang.org/learn has additional resources as well once you have gotten through those.

My point is the different levels of just working are subjective, not objective. I personally have spent far more time fixing bugs or just reinstalling ubuntu systems then I have over the same period for Arch systems. So many of my ubuntu installs just ended up breaking after a while where I have had the same Arch install on systems for 5+ years now. Could never get a Ubuntu system to last more then a year.

Everyone has different stories about the different OSs. It is all subjective.

There is no perfect OS that just works for everyone. They are all software so they all have bugs. People how say an OS just works have never hit those bugs or have gotten used to fixing/working around or flat out ignoring them.

This is true of all OSs, including Windows, Linux and MacOS. They are all differently buggy messes.

Linux is the buggy mess that works best for me though.

Realtime is important on fully fledged workstations where timing is very important. Which is the case for a lot of professional audio workloads. Linux is now another option for people in that space.

Not sure Linux can run on microcontrollers. Those tend to not be so powerful and run simple OSs if they have any OS at all. Though this might help the embedded world a bit increasing the number of things you can do with things that have full system on chips (like the Raspberry pi).

By clear receiver it means there is only one function a name can point to. For instance you cannot have:

struct Foo;
impl Foo {
    pub fn foo(self) {}
    pub fn foo(&self) {}
}

error[E0592]: duplicate definitions with name `foo`
 --> src/lib.rs:5:5
  |
4 |     pub fn foo(self) {}
  |     ---------------- other definition for `foo`
5 |     pub fn foo(&self) {}
  |     ^^^^^^^^^^^^^^^^^ duplicate definitions for `foo`

Which means it is easy to see what the function requires when you call it. It will either have self or &self and when it is &self it can auto reference the value for you. This is unlike C and C++ where you can overload a function definition with multiple different signatures and bodies and thus the signature is important to know to know which function to actually call.

Yes rust has a operator for dereferencing as well (the *). This can be used to copy a value out of a reference for simple types the implement Copy at least.

On quite a few terminals, alacritty is one, you can hold SHIFT while dragging to select things on the screen to copy from. This suppresses sending mouse events so works when you have editors, tmux or other things that can interpret mouse clicks. It does mean that it is just copying chars from the terminal so ends up copying everything on the screen. Unlike the copying baked into things like tmux of editors.