I start: the most important thing is not the desktop, it’s the package manager.
That after getting used to Linux I will hate to be forced to use less free operating systems.
This so much. I absolutely cannot stand Windows anymore.
Windows is so bad
The 1:1 windows:Linux replacement is just a means to keep you on Windows. Once you learn Linux, you’ll come to understand how much of a farce it is and how it’s designed to keep you away
I guess the main things would be:
- As a beginner, don’t bother trying to dual boot – If you still need a Windows box, get some cheap hardware to do your Linux work on. It’s too easy to screw up both systems otherwise.
- Don’t get too hung up on a specific distro, the better you are at dealing with different configurations, the better prepared you will be for whatever comes. Once you’ve gotten one set up, don’t be afraid to just try a different one.
I did the opposite, have always dual booted my laptops and had win on my PC until quite recently now that I’m comfortable enough not to need a safety net anymore
If you have the space for a spare I much prefer hot swapping hard drives. it’s a little physical inconvenience but much harder to screw anything up. plus, full disk encryption is still an option
Trying not to make it windows.
There’s a lot of conveniences that Windows comes default with.
When I switched to Linux, my immediate goal was to find alternatives for EVERYTHING. That lead to being disappointed by a lot.
Understanding Linux and also recognizing there’s a lot of shit I don’t need (that windows was giving me for the sake of VALUE) was a game changer.
Understanding Linux and also recognizing there’s a lot of shit I don’t need (that windows was giving me for the sake of VALUE) was a game changer.
This 100%! After using Linux for the past few years I’ve realized a lot of the crap windows has by default is stuffed in there to have something to market.
Nowadays there’s a lot of good alternatives for everything, including windows hello for any password prompt
I learned to never settle. If you don’t like the default workflow of Gnome, try some extensions, or even a different DE. Same with Package Managers. If you don’t like the syntax, make an alias. Don’t just “deal with it”. Windows has brainwashed people into thinking that there is only one way to do a thing.
This is kinda funny to me because I hadn’t realized how terrible the Windows workflow was for me until Gnome 3 came out.
Ever since, while I’ll use extensions for stuff like alphabetical app grid and Caffeine, I never do anything that changes the Gnome workflow. It’s not for everyone, but it absolutely is for me.
Its why I always find it funny when people complain about changes to the start bar, because surely there isnt a bunch of 3rd party options in existance that change it, and can mimic 7’s start bar.
I have heard that shell replacements are often very buggy on Windows.
Ive been using classic(then open) shell since moving off of 7 for consistency. for the most part, there haven’t been any serious bugs that im aware of. Because the app works between windows versions, start bar for me at least has been pretty much consistent since windows 7 existed, and the stuff id adjust to would be changes in some apps (e.g control panel > settings) that happened overtime.
The problem of some users is they want the vanilla experience to be what they want when there are options to not make something vanilla. Similar to debates on linux distros on whether you want a very specific UI design vs having a distro that is personalizable and customizable based on preference.
It was free, I could not afford a Sun workstation and Minix had problems, so when this Finnish guy wrote in Usenet that he was working on a free kernel/OS, it was cool!
386BSD was a thing back then too, but there was the AT&T lawsuit that scared everyone away. That gave Linux an opportunity.
Distrobox exists, so one is not bound to use a specific distro just because it packages some of the apps/binaries they require.
Installed distrobox on NixOS because I was worried being limited to only nixpkgs and have not touched it once lol
Same goes for the windows VM except for the time I needed to run excel macros for work
I did on my Nix, there was a package in Nixpkgs that was outdated, so I had the opportunity to use distrobox for that, at leqst temporarily until they update the package.
Am I reading the readme correctly in that I can run apt-get within distrobox on Fedora, and not be limited to dnf packages?
You can install Distrobox on Fedora (or any of the distros that support it), create a Debian distrobox on your Fedora install, and within the Debian distrobox you can use
apt-getto install whichever Debian package you like. Or…, you could make an Arch distrobox and even install stuff from the AUR. Or really any package from any of your favorite distros as long as it’s supported.Awesome! And it’ll be segregated from the base system and from other containers, like toolbox installs are?
And it’ll be segregated from the base system and from other containers, like toolbox installs are?
Exactly. It’s even possible to segregate it beyond what Toolbx has been able to do (at least since the last time I checked) in that you can define another folder/directory as your HOME directory within the distrobox.
Amazing!! Yup. Looks like this is getting installed on my Fedora tonight. Thanks!!
Glad to be of help 💙 ! Feel free to inquire if you so desire 😉 .
I appreciate that!
Yes!
Unmounting removable drives after writing to then is crucially more important than on Windows
It’s pretty important on Windows too, though. Always “eject” or “safely remove hardware” before unplugging!
How so?
On Windows, I often simply took out the USB drive without “safely removing” it. The data was there 99% of the time. On Linux, if I’m not mistaken, unmounting the drive before disconnecting is what actually writes data to it.
I don’t think Linux literally waits for you to unmount the drive before it decides to write to it. It looks like that because the buffering is completely hidden from the user.
For example say you want to transfer a few GB from your SSD to a slow USB drive. Let’s say:
- it takes about half a minute to read the data from the SSD
- it takes ten minutes to write it to the USB
- the data fits in the spare room you have in RAM at the moment
In this scenario, the kernel will take half a minute to read the data into the RAM and then report that the file transfer is complete. Whatever program is being used will also report to the user that the transfer is complete. The kernel should have already started writing to the drive as soon as the data started being read into the RAM, so it should take another nine and a half minutes to complete the transfer in the background.
So if you unmount at that point, you will have to wait nine and a half minutes. But if you leave it running and try to unmount ten minutes later it should be close to instant. That’s because the kernel kept on writing in the background and was not waiting for you to unmount the drive in order to commit the writes.
I’m not sure but I think on Windows the file manager is aware of the buffering so this doesn’t happen, at least not for so long. But I think you can still end up with corrupted files if you don’t safely remove it.
Or just
syncbefore pulling it.I do not think this is the case. You can disable on GNOME Disks active disk write caching for removable storages, exactly the same way as on Windows.
Also, Thunar File Manager has an option to partially write files when copying/moving and when moving, only remove the file from source directory when the copy is successful. I find it remarkable against data corruption for large file transfers.
Yeah, but you just describe 2 features on specific apps that don’t need to be enabled by default.
I mean, even the SATA over UAS is a pain with Linux, since the new implementation sacrifices SMART data for faster RW speeds on Linux, and you have to fallback manually on the old driver to read SMART data on external HDDs. On Windows, you just use CrystalDiskMark and it works.
Linux needs you to do a little work here and there for such things. I do not really eject everything safely on Linux. The feature on Thunar is handy.
Rasberry Pi or other NUC is a great way to begin.
By the time you’ve dressed out an Rpi to be halfway usable, you’ve spent about as much as a decent NUC. And all you have to show for it is a slow-as-mud sd card, hardly any video acceleration, a USB stack that only crashes sometimes, a busy OOM killer, and no software.
Get an N95 based nuc. A Beelink with 8/256 runs about $150, and it just works. (Well, you might need pcie_aspm=off).
yeah, RPI is just ‘cookbooked’ due to fixed hw
When you’re just trying to get work done: pick a solid, well-tested high-profile distribution like Fedora, Pop!_OS, or Debian (or Ubuntu). Don’t look for the most beautiful, or most up-to-date, or most light-weight (e.g. low CPU usage, RAM, etc.). Don’t distro hop just to see what you’re missing.
Of course, do those things if you want to mess around, have fun, or learn! But not when you’re trying to get work done.
Is Pop!_OS really that popular? I started using Linux about 10 years ago and it wasn’t around then, so I never tried it in my distro hopping days. I see it’s developed by System76 so I can see why you’d choose it on their hardware, but is there any point doing that on other hardware?
The System76 engineers are culturally very aligned with the core values of freedom of choice, customization, etc. They build software with the larger ecosystem in mind, and in fact, I’ve never seen them build something only for their own hardware (even things that could have been just for their own hardware, like the system76 power management system, has extensibility built in).
That said, they also balance this freedom with a set of “opinionated” good choices that they test and support. If you care a lot about stability, it’s easy to go along with the “happy path” and get a solid, up-to-date system delivered frequently. Every time they upgrade new features or kernel, they go through a systematic quality assurance process on multiple machines–including machines not of their own brand. (I’ve contributed software/PRs to their codebase, and they’ve always sent it through a code review and QA process).
Idk, it seems to be picking up steam. It’s what I use unless I’m trying to use something super lightweight.
For me it has the stability of Ubuntu without having to use Ubuntu.
Haven’t tried Debian yet though.
I’m cirious about what you dislike about Ubuntu?
There’s a small amount of telemetry going on.
Also, Pop_OS makes running an Nvidia GPU less painful.
Snaps are basically Ubuntu’s private app store, and flatpaks (the supported method of app distribution by almost every other distro) are not supported; there’s no tiling WM built-in for large monitors; the kernel is not kept up to date (i.e. improved hardware coverage and support); some things like streaming with OBS studio and Steam don’t work out of the box (this may have changed, but it was the case for me about a year ago).
Interesting, thanks. I had a feeling snaps would be in the list!
I wish I’d known how much of a pain in the ass having an NVIDIA card would be. I would have gotten a different computer.
Same. I bought my GPU at like 170% of its MSRP. I regret it now, should have went the amd way
That I shouldn’t care as much about installed packages I no longer use. Sure, going through installed programs and cleaning up from time to time is ok, but no need to panic if something sticks around.
Especially when I installed something manually needing dependencies for programming, I tend to write down names of installed packages and then managing it manually, because I wasn’t yet aware what their names mean. Now using same OS for over a year, heavly testing stuff, having multiple desktop enviroments and not cleaning it up my system partition is taking less than 30GB, compared to 1TB disk it’s nothing.
To start sooner.
TIL there’s tab completion lol
Tab completion, history, history search, you name it
Always put your filesystems in an LVM volume (and in general, partition disks with LVM rather than partition tables)! You never know when you might need to combine multiple disks, make a snapshot, add redundancy, or transfer to another disk without unmounting. But it’s very difficult to format a block device as LVM once you can’t erase its contents.
Make your /boot partition at least 500MiB.
Leave at least 1GiB of free space at the beginning of every disk. You never know when you might need to add EFI and boot partitions to that disk. And again, it’s very difficult to do after the fact.
