Not sure there will be a big change there, because they are already powerful enough for most common tasks since several years now. And everyone owns at least one phone or tablet already. So I don’t think that number is going to rise significantly anymore. Those people who are OK with using a phone/tablet for everything probably already do so right now. Maybe if living conditions for the non-super-rich become worse and people look for more affordable computing devices. But even then, older devices which can run Linux desktops for example are already dirt-cheap. I just don’t think that the UI/UX of phones or tablets is on par with desktops or notebooks running a regular desktop OS when using a big screen. Those UIs are primarily made for touch and for smaller screens. Trying to do everything with just one UI paradigm just leads to Windows 8 ugliness.

Desktop Linux’ marketshare is going to steadily increase, but as time progresses, so will the speed of that increase. Linux was at or below 1% for a really long time but within the last 5 years or so it jumped to ~5%. As this not only means more users, but also more attention and developers, this will of course snowball. The end of Win10 will also give a bump. And if the enshittification of Windows continues (it probably will) and if US-based companies are becoming a red flag for non-US-customers (will probably also happen) then it will snowball even faster.

• Tor browser for anonymous/private regular browsing (without logging into personally-identifiable accounts)
• Vanadium (GrapheneOS’ Chromium-based browser, maybe it’s usable on non-GrapheneOS as well?) in combination with a good crap-blocking DNS server
• Brave is decent but has some bad default settings, can probably be configured to behave well (similar to regular Firefox)
• Firefox + forks are generally not that great (at least on Android?) because their sandboxing capabilities (and maybe other security features) are weaker compared to those of Chromium-based browsers. See also: https://grapheneos.org/usage#web-browsing
• Proprietary browsers like Chrome, Edge, Opera, and so on all contain loads of on-by-default-spyware and should never be used

An easy analogy that common users can understand is e-mail. E-Mail is also decentralized, everyone has an e-mail address but everyone uses a different e-mail host (the domain name after the “@”). So e.g. “john.doe@gmail.com” has an account at gmail.com but “jane.doe@mailbox.org” has an account at mailbox.org. Both are completely different, yet they can communicate with each other. There’s not one company controlling or storing every single e-mail account or inbox. It’s spread out and everyone can choose the mail provider they like or trust the most.

Then you use that as a bridge to explain Lemmy, or Mastodon, or other Fediverse social media platforms. And remind the listener that single companies having full control over everyone’s accounts is generally bad and opens the door for all sorts of abuse and manipulation or arbitrariness.

Sure.

• https://www.scss.tcd.ie/doug.leith/apple_google.pdf
• https://www.scss.tcd.ie/Doug.Leith/Android_privacy_report.pdf … and when you use for example Samsung, Xiaomi, Huawei made mobile OSses, it’s even worse of course, because then their own telemetry comes on top of that from Google as well. They want to have pieces of the cake too.
• https://www.kuketz-blog.de/google-play-services-die-ueberwachungswanze-von-google/ (you might have to translate that page. That guy is a well-known privacy and data protection researcher, white hat hacker and also a data protection official in a federal state in Germany. He’s very objective)

Just for reference, this is what the Google Play services app transmits roughly every 20 minutes to Google if it has network access:

Phone #
SIM #
IMEI (world-wide unique device ID)
S/N of your device
WIFI MAC address
Android ID
Mail Address of your logged in Google account
IP address

And that is when you have disabled ALL telemetry in ALL of the options, even the most hidden ones. So this is the minimum amount this app is always gathering from every Android user using the Google Play services app, no matter what you selected. Other Google apps (like the Play store app) could then contain additional telemetry on top, this is just the common base of all Google proprietary apps. Or the minimum amount of privacy violations you get when using proprietary Google apps on your phone, no matter what.

If you use GrapheneOS, I’d recommend not installing/using ANY Google apps at all (not even Play store or Play services). To get apps, you should use (roughly in this order of priority): 1.) GrapheneOS’s app store for the built-in apps 2.) Accrescent app store (has several good open source apps, is intended to be more secure than F-Droid) 3.) Obtainium (for getting open source apps directly from their source repos) or if you really can’t get into Obtainium, use F-Droid instead 4.) Aurora Store (for getting apps from the Google Play store without sending too much data to Google. Only do this if there is no open source app available for doing the same thing).

To fully mitigate the removal of the Play services app, you also should probably install/configure something like ntfy with UnifiedPush to get battery efficient push notifications and ideally use apps which also use that, e.g. the Molly fork instead of Signal. It’s quite easy to do, just something to be aware of. Otherwise your battery drain might be a bit higher. Then you’re also independent from Google’s push notification infrastructure. But you need a UnifiedPush server to go along with it, either self-hosted or use a public one. There are some privacy friendly ones public ones out there.

Haven’t read anything bad about Tuta so I guess it’s fine. Other good ones are Proton, mailbox.org or posteo.de. Anything that’s not by Google, Microsoft, and so on.

I use several, depending on use case:

• Tor Browser for general and anonymous web browsing (e.g. reading news, looking up stuff, and so on)
• Mullvad Browser as a clear web alternative for general use
• Librewolf for generally logging into sites with personally identifiable accounts (e.g. to buy stuff)
• Ungoogled Chromium for those few sites which only work with a Chromium-based browser, or other specific cases
• On Android (GrapheneOS): Tor Browser and Vanadium

All regular browsers have some hardening applied and uBlock Origin installed.

Since you only mentioned 25% gaming, I’d recommend against a gaming-centric distro like Bazzite. Instead, use a generalist desktop distro. Since you mentioned that you’re rather new-ish, I’d recommend against Arch-based distros like CachyOS. Instead, check out e.g. Fedora, Mint, OpenSuSE. (Probably in that order of priority) These aren’t hard recommendations, so you can do whatever and probably be fine either way, but it still doesn’t fit that well.

In theory, the law applies to everyone. In practice, it doesn’t apply to rich entities.

Well, “nice” to have that feature, but you still shouldn’t use Discord, or at least limit your time on it as much as possible. Remember, when you use it, you’re part of the network effect that makes Discord big. You have to resist that. Take a look how Discord compares to pretty much any other messenger or communication tool in existence: https://www.messenger-matrix.de/messenger-matrix-en.html Avoid it whenever possible, get your friends to leave it and weaken its network effect.

So, some of the drawbacks (there’s probably more):

• Discord has weak security (see URL above)
• Discord has non-existent privacy
• Discord has an incredibly vague privacy policy which means they do what they want. Even companies with strong privacy policies screw users over routinely. Guess what companies do who don’t even care about good privacy policies. They even weakened it further a while back because they need to train their new AI features on your data as well, and probably even their weak privacy policies were in the way before. Well, good thing that the users agreed that they can change it at any time for any reason and be fine with it.
• You grant all rights of everything you write, say, share or do on Discord to Discord, and everything you type, say, upload or share is being processed by their servers
• Discord tracks what you’re typing before sending it
• Discord suspended accounts and required even more user data for “verification”, such as telephone numbers which is completely unnecessary except for tracking and data selling purposes
• Discord shares chat logs with law enforcement (and they can share everything because they’re collecting everything)
• The Discord client app tracks what programs you have running so it can for example display what games you’re currently running. That means it contains a process logger. It has to scan every running application and then finds games out of those, and then you have to hope that only the game-specific bits are uploaded to their servers. Maybe that is the case, but can you trust them, and also to never change that? No.

If you have to use it:

• Never use their desktop app, always use the web version from a secondary browser (web apps running in the browser have much less rights than locally running applications), and even then limit what the site can access to the least amount of stuff necessary. If you never use your mic or camera then block it in the browser settings. Don’t trust Discord’s own mute setting (this also applies to other proprietary software).
• Use a fake e-mail alias / telephone number when creating your account, generally give them the least amount of data possible. Opt out of any options or features which are tied to you exposing more data to them
• Don’t give them additional money e.g. for their premium stuff (you already pay with data they gather from you)
• Block at least these API endpoints which are purely used for tracking purposes (there may be more though, and they might change) [easy to do with uBlock Origin for example]:

https://discord.com/api/v*/science
https://discord.com/api/v*/channels/*/typing
https://discord.com/api/v*/track*

• You can also block these related hosts safely without impairing Discord’s main functions (again there may be more):

crash.discordapp.com
status.discordapp.com
b.stats.paypal.com
app.adjust.com
client-analytics.braintreegateway.com

Yeah, I also don’t like such general laziness. It’s also not just limited to switching to Linux, it’s kind of the same with switching to anything that’s better but slightly(!) more inconvenient than what you’re used to. Well, you can’t make or be part of some progress unless you’re willing to sometimes get off your comfy couch and do something you’ve never done before. Like switching to Linux. Like stopping eating meat. Like stopping supporting certain evil companies. Like going to vote for a non-retarded option. Like voting with your wallet for the products you use/buy and also NOT use/buy. If everyone would do it, the world would be a different (better) place. But still too few are doing it. Because it’s slightly less convenient. And that would be so damn hard to change. Oh man would that be hard. Not.

Distro hopping is fairly normal if you’re still relatively new to Linux, I guess you do it less as time goes on, because you’ll have a better idea of whether or not a specific distro is appealing to you or not. To be able to even judge that you have to try out some distros for yourself, of course, so you need to do some distro hopping in order to tell what “direction” of distro is best for you. Sure you can read about it or watch videos but it’s never the same as actually running it for yourself.

The question is kind of impossible to answer because the two are so different. It doesn’t make sense to compare Linux to a version of Windows.

Also:

One side (myself included) is usually disgusted at Windows for being so bloated, full of spyware and dark patterns, closed, untrustworthy, fraudulent and inefficient. So personally, I’d rate Linux to be as good as a non-existing future version of Windows that’s never going to appear.

While the other side (most “average users”) are usually disgusted at Linux because Fortnite, Photoshop and that random stupid thing they bought at some store don’t run on it. As stupid as it sounds but that is usually enough reason for an average user to not like Linux. Also, he’d have to install it himself because it’s not preinstalled. Also a major hurdle for that kind of user. Unfortunately, the majority of users. Users like that probably rate Linux as bad as something like MS-DOS or Win 3.x because they feel that Linux is limiting them, but at the same time don’t want to change anything about their software choices.

The main problem is that common users are usually tied to specific proprietary software (or specific formats which can only be opened by specific proprietary software) which ties them to Windows, and anything that doesn’t run that software cannot be an alternative for them. That’s probably also the reason why MacOS isn’t more popular because it also can’t run everything, but it’s still better than Windows. So unless those users change their habits and the software they use, they will always be shackled to Windows and remain on that sinking ship until they’re literally beneath the ocean, because they never realize a sinking ship.

It’s important though to not fall into the trap of creating false balance, i.e. giving the same weight to a false or harmful statement than to a truthful or good idea, in the name of “fairness” or “objectivity”. Also, conservatives tend to shift to the right currently.

netstat is kind of deprecated, ss is more modern (from the iproute2 package) and uses very similar parameters.

• awk
• the (usually rust-based) coreutils “alternatives” like bat, fd, eza, procs
• trash-put (rm with trash integration. But beware that it also operates on directories by default, which rm only does with -r. There should be an option to change that behavior but there isn’t. Don’t alias rm to this)
• wl-copy/paste (or the older one for X11, ‘xclip’ IIRC. Enables you to do stuff like “cat image.jpg | wl-copy” to copy it to the clipboard. Best alias it to something shorter)
• xdg-open (open the file using your associated program for that file type. Alias to “o” or so)
• pass (awesome password manager, when you have a GPG key pair. Even better in combination with e.g. wofi)
• notify-send (to send GUI notifications from shell scripts)
• ledger (plain-text accounting software. If you use Emacs you should take a look at this as it’s written by an Emacs dev, and has good integration of course)
• nc
• nohup

• To learn Arch, install it from scratch (without archinstall), it’ll force you to read the Wiki and learn a lot of necessary commands in the process. After the installation, just keep using it. Using a Linux distro full-time as the only installed OS is the best way to keep at it and truly learn it over time. There’s no magic bullet here. Just keep using it and solving problems or issues as you go, learning more and more stuff as you go. If you need other OSses as well, run those in a VM. I don’t recommend dual-boot setups.

• Don’t blindly copy/paste commands you don’t understand. Always try to understand them first. Some commands can be very disruptive or even destroy your configuration. If you don’t understand it or are able to adapt it so that it fits to your particular configuration or system, you can EASILY damage a configuration, or even make your system unusable. Also, some people like trolling other people and deliberately share harmful commands. Generally, test potentially destructive commands or complex commandlines before actually running them.

• Document major config changes that you do. This is useful because you’ll be able to undo certain changes or even replicate your current system configuration fast when you change distros or have to reinstall in the future. For example my current Arch-based setup is fully documented in form of an almost-directly executable shell script. It does require some interactions but very little. If I ever have to reinstall this system, or upgrade my hardware, it can be done insanely fast and it’ll have the exact same configuration. This goes from basic partitioning and encryption all the way up to dotfiles and individual program configurations.

• Don’t feel the need to learn hard/advanced tools like Vim or Emacs unless you really think you’re getting an advantage from that and aren’t hesitant to put in the time and effort to learn them. Most people don’t need to use them. They’re amazing tools but you need to be prepared to lose quite a lot of time to learn them before you can become productive with them, and this might not be a tradeoff that’s useful for every single user. You can also get away with much simpler tools, like nano (as a console-based editor) or whatever programmer’s text editor you want.
• Similarly, whether a pure WM or compositor plus assorted tools compared to a full desktop environment is worth it for you or not, is up to you. There’s no wrong or right answer here. I’ve tried out pretty much everything and these days use KDE Plasma because I like the consistency and integrations and dislike having different, inconsistent stand-alone tools for panels, menus, notifications, wallpaper, file manager and so on. But again, there’s no wrong or right answer here. Just what makes more sense for you. It’s worth learning how to be able to configure and use a minimalistic setup, for sure. So trying it out doesn’t hurt and increases your knowledge overall. In general, in the Linux world it’s good to always know enough to not be screwed once some component suddenly doesn’t work anymore. For example, a competent Linux user should be able to deal with (temporarily) not having a GUI and fixing his system via commandline.

• A minimalistic, DIY distro like Arch can be amazing to learn everything, if you want to do that at least. If you just want a working desktop system with as little effort as possible, then don’t do that. But if you intend to learn every detail, then a distro like Arch is better suited for that goal than a “bloated”, fully pre-configured distro like Mint or Ubuntu is. Because Arch is much simpler on a technical level than those are. It’s much easier to understand e.g. the relatively simple package building process on Arch than it is on Debian/Ubuntu-based distros. But this “simple” explicitly refers to technical simplicity or minimalism. Most users expect something else when they hear “something is simple”. Arch is not simple as in beginner-friendly, but it is simple in terms of technical complexity, which is why many advanced users and tinkerers like it because it doesn’t stand in their way. It also means though that you HAVE to learn many things, e.g. how to configure a firewall, because it doesn’t come with any preinstalled by default. With Arch, the admin is supposed to know about everything and configure every component himself, at least on a somewhat basic level.

• If you want to go to even more details, you could also try out a source-based distro like Gentoo or Crux, which can also be a great learning experience, but it’s even more details regarding compilations of each package, dependencies, compile-time options, etc. you have to deal with than with a minimalistic binary-package-based distro like Arch, so whether that’s useful for you or not is up to you of course.
• While we’re at it: LFS (Linux from Scratch) is not a distro per se, it’s a guide on how to build your own distribution from scratch. It’s VERY time intensive and not recommended unless you truly want to learn how to build a complete distribution from scratch, or maybe start your own distro some day which isn’t based on another existing distro.

This is a good opportunity to remind people that they should donate a bit to some of the open source projects which they find helpful or important. Liberapay is a nice platform for doing so.

Well, it might seem that way sometimes. But in the end, what’s different to religion is that this is all rooted in facts. Facts which are quite abstract, so not everyone gets them and even those who do get them sometimes wonder whether it’s important or not sometimes. The thing is, Linux is at its core a neutral, open and free operating system, and it’s basically the only one which is advanced or mature enough to be a real competitor to let’s say Windows or MacOS. Of course it’s more than a competitor on the server, it’s basically the only relevant server operating system (Windows Server has a niche in application servers within a MS intranet domain, or to control Windows clients via policies, that’s about it, and MacOS server is already long dead I think). Of course, some of Linux’ success is because those same companies also contribute a lot to the development of Linux, because they need it for themselves as well. But that’s just one more thing which makes Linux a very unique thing. It’s like a neutral baseline for an operating system. Like a very capable OS core that everyone works on, even the competition works on it, because they also rely on it.

That it’s open source and transparent and that anyone can use it or improve it or change it or whatever makes it special, because it’s not a commercial black-box product where you just consume it as-is and have zero rights whatsoever to do or change anything about it. That’s actually incredibly special in today’s commercialized landscape. Its open nature also means it can never die, only grow. And because it’s a proven good system which is also so very different compared to established desktop OSses, it can happen that its users or fans can seem somewhat religious towards it. But, again, compared to religion, religion is based on pure belief (otherwise it would be called fact). There’s nothing religious about Linux or open source software. It’s simply a special operating system, and not in a bad way at all. And closely related to it is, of course, the whole free/open source software movement. Which every user, even those of closed operating systems, can and do benefit from.

And since today’s commercial software continues growing more and more user hostile (ads, spying, bloat, dark patterns, high prices/software rental models), it’s getting increasingly important to have at least the option of a true alternative. Even users who absolutely hate Linux and open source software should be glad that alternatives do exist, so that once the food they are being fed by Microsoft and so on doesn’t taste good anymore, they at least have an option to switch to something else entirely.