Ive always told them that Linux is:
Please don’t…
This reminds me of vegan people.
My favorite part is that X.org is able to render 16K resolution and has been able very long time.
So it’s 16K ready, which means. That we need need to pull out our linux GUNS ?!
1.16K rap songs about linux in 16K desktop.
2. We tell people that all medicine is actually just a derivative of dpkg or a dependancy of it
3. We say that the only method to safely and succesfully remove all CO2 from the atmosphere can only be removed by a linux monolithic kernel
Yes but if people are actually pissed off because Win 10/11 legitimately sucks you could always do a gentle nudge without being preachy.
I’d say “We don’t have to suffer this each time. There is a better way to do things but it needs a bit of time because there is a learning curve that is worth it. ”
I have converted 2 of my workplace to Linux because the use case is simple enough. I am tempted to convert another but the company is big enough and their IT is a dedicated Windows guy and may need to relearn the whole windows thing to move to Linux.
Obviously you are free to do what you want. I don’t have much more to contribute to this thread so I’ll see myself out.
That’s something that you CAN do, but is it something that you SHOULD do?
I had already been on the Internet for years before America Online happened, and I saw the IQ of the community fall through the floor in real time when the floodgates opened. Do we want the same thing to happen to Linux?
The small linux problem thread - #4809 by GTwannabe is overwhelmingly full of intelligent questions and the vast majority of posters put in real effort before asking for help. Imagine that changing to a never-ending stream of “it dont work”-type questions, of which 90%+ are ultimately resolved by simply getting the user to apply ignored updates and reboot their machine. Is that something we really want?
Popularity is a very poor indicator of value. Linux has proven itself to be valuable despite not being popular. Indeed, perhaps the only way an OS (or distro) can be of value is for it to be free of the soul-crushing inertia created by hordes of clueless normies?
So, before getting too carried away trying to greatly increase the size of the Linux user-base, pause and give some thought to the question of whether that is actually a good idea. What will dropping millions of metric tons of dead weight onto the OS do to it?
Dude i agree with everything you’ve stated and thank you. I uploaded some 4K linux content, but barely get any views. Not because im so interested in the popularity but have u seen how fast my content has declined?
I’ve gone from this
To this
And this
TBH, if you want normies to use Linux, you need to make Linux not Linux so to speak. Something closer to Android, but still feel linux-y.
I would imagine something based on Fedora Silverblue or OpenSUSE MicroOS, both running on BTRFS to snapshot the OS and to only run Flatpaks and Appimages. Or a completely snappified Ubuntu, running on ZFS. That way, the core OS would still be Linux, but programs would be confined in their own somewhat secure enclave and hopefully never require root permissions. Then we need permission options for each program, like Android has. For example, permission to use the camera, microphone, storage, and if it has access to storage, which folders only, not to access everything, permission to communicate with other program, permission to use the Internet etc.
The Linux desktop needs an immutable system. Both Silverblue and MicroOS are immutable, they give you transactional updates and if updates fail, at reboot, it will automatically come back to its last working state by going to the automatic snapshot that was taken before the update, or if that fails, the snapshot before it and so on. I believe Snappy Ubuntu also does that, but I don’t know its status. Ubuntu desktop has a ZFS experimental install and I read online that Canonical probably plans to make it default at some point, meaning that the system can be immutable too, thanks to ZFS snapshots. Add snappy to the mix and you get a pretty powerful desktop.
Obviously, all 3 of them can run OCI containers to expand on their software selection, like say, if people want to run Element (Matrix client) on their desktop, or host bitwarden_rs or booksonic on their desktop. I personally don’t mind appimages at all, flatpaks are ok too, but snappy sucks because its dependency on systemd, because of automatic updates, because of its centralized software repository and because snaps create additional mounts on the system. If Canonical could address those issues, I’d be fine with snaps too.
Obviously, being more of a Linux dude, I prefer classic package managers if I can help it, but for normies, I would recommend them an immutable system and sandboxed / contanerized software any day.
That way, Linux will “just work” for them, so they don’t have to worry about the spooky terminal. Silverblue seems the closest to being possible to use. MicroOS is still sketchy from what I read, and I have no idea about Snappy Ubuntu, because Canonical shifted their focus on servers and IoT. Snappy Ubuntu Core probably works amazing for the companies that make use of it, but I don’t think it is even remotely close to being able to be used as a desktop.
You don’t have to centralize Linux in order to make the experience good. You just need software to be packaged for Flatpaks and Snaps more. I don’t really understand what’s difficult about that, but I never packaged software. But in theory, it should be easier, because you are basically doing a chroot environment on steroids, with no dependency hell to screw with your package.
Valve could probably launch a Flatpak version of the Steam store for Linux game developers to publish their apps, and to keep their stupid idea of DRM without introducing rootkits and spyware on a system, make the flatpaks require communication with the Flatpak Steam Store in order to launch the game.
But again, my opinion on this matter is that you don’t want to “convince” people to switch, you want people to already want to switch. You can only do that by offering a better experience for them.
If Android didn’t had such a dumb design, requiring basically custom OS for each SoC, Android could have become a serious alternative to desktops. But you can use a desktop for 10 years and still receive updates and there are 20+ years old PCs that can still be used today by running Linux on them. Android gets what, like 5 years of updates at most if you’re using a Google branded device, and like 2 years on average at most? Sure, software won’t suddenly stop working, but they just might one day, while security becomes a big concern. Linux can more easily tackle this artificial deprecation of hardware on both PCs and phones. Google needs to get their stuff together, maybe FusciaOS will be similar to how Windows and Linux roll updates, which could be a viable platform for mobile, desktop and IoT. But it feels like Google is kinda late to the game.
Linux needs better ways to make a switch possible.
Such as, a website with ram or nvram access. To store an installation ISO so someone doesn’t need the usb installer thingie. That’s just one idea, but having such a server or something up and running with viable https access etc. instead of the sketchy things, it should be possible to get done somehow
It used to be that ubuntu (probably other distros too) had cds that you could get for just the cost of shipping. I know I picked one up for free at ohio linux fest once. I think the general problem with providing physical media was that the media gets out of date and so it increases the chance of hardware incompatibility during install
There is no step 2 through X
Normies can figure out all sorts of ways around enterprise firewalls, etc. if there’s motivation to do so. They generally aren’t stupid. If the motivation is there they can figure things out.
The problem with linux is that there’s always some sort of hassle with some sort of software that is critical for the end user.
The absolute worst thing you can do imho is try and switch someone over before they are ready and/or before all their critical stuff runs on it acceptably, as all you’ll achieve if they don’t figure out how to deal with it is leave a bad taste in their mouth and taint the experience for them.
The only way is to get them young. If schools and colleges have Linux in their class roooms and labs normies would grow up with it and be comfortable with it.
This will be a thing eventually.
Apple figured this out decades ago with their education discounts.
But yeah, linux is used heavily in universities, eventually that level of knowledge will trickle down to secondary schools, etc.
The Raspberry Pi is going to do massive things for this. The Pi400 is almost acceptable as a genuine low cost desktop computer (performance is not QUITE fast enough imho). You could certainly use a fleet of them for a class room in a pinch, but I really think the pi 5 will be a genuine desktop capable machine with few compromises for education.
It is like losing weight. When you are ready you are ready. When you go on fad diets, you go off and gain more weight than before. It might not be exactly this. But I have tried linux over and over. Everytime I go back to windows. Why? I run into major issues or simple annoyances. Windows just works for the average consumer.
Compatibility is way better with windows. Just for instance, I had to edit configs to get my manually set audio output device to be default. In windows it is as simple as going into control panel. Linux, you have to google how to do it and then risk breaking stuff in the process of doing something as simple as setting the default audio device.
Another issue I have is display resetting their orientation and order after waking from sleep. I fixed it by figuring out the order on the gpu itself. This is not an issue for mac or windows. Simple stuff like this means linux is not ready for joe average consumer.
Gaming is better with proton, but still not prime time ready. And to have to do something as complicated as hw passthrough for getting a more windows experience defeats the purpose of leaving windows anyways. A compatibility layer is buggy still (such as proton). My game instances often freeze after alt tabbing or lose audio altogether. So a restart of the game is required.
Libre office almost always imports docx files incorrectly. If you start a fresh document in libre, it has issues with formatting in actual word.
It is all this kind of stuff that means, linux has a long way to go, before I let my mother who is 74 use it.
Hell runescape native app for linux is trash. so i have to use proton and it has issues in and of itself. The native gaming experience for runescape leaves me with a less than ideal experience. I have to restart my game (something I don’t have to do in windows), because it freezes quite regularly. Which leads me to the idea that linux doesn’t have support for AAA developers or indie for that matter. They focus their developing on mac and windows.
With all this said. I realize none of what I said above helps in getting people to switch to linux. Other than the fact that people don’t like dealing with new stuff that is more complex than what they are used to. New things scare the susie q, and less compatibility also scares joe blow.
To get someone to use linux, take older hardware that either runs really slow or doesn’t support up to date windows and tell them linux still supports that old stuff. Tell them that you are more secure using up to date os’ and software. That is about it.
Also tell people that linux runs smoothly on a 10 year old pc vs windows 11.
But the best thing is to not force it down their throats. Try to convey that linux is free and could be tried out at not additional cost. (even if it is not true, cause time googling stuff is time not well spent). I have completed a week on popos this time, normally I make it 2 days before rage quitting.
Also tell them if set up properly it can do 99% of what windows can.
The Linux Zelot community lines up very nicely between vegans, jehova’s witnesses and ex-smokers.
I’ll be doing my part soon, lmao. First PCs for 2 kids are going to run Linux. I’m taking this too seriously. The only reason for that though is that I’m a cheapskate (budget PC builds, will make them Windows 11 compatible, but I want to see if they’re going to work with Linux first, because it’d be easier for me to manage them in a pinch - and no need to buy a Windows license).
I don’t know why wait that long, the Pi 4 8gb is my main PC already (writing from it) and has been for the past ~half a year.
Maybe it’s just my bias, but I’ve seen Windows misbehave a lot more times than Linux. Given, the number of Windows to Linux users where I worked was around 5:1, yet the Linux users never complained about Linux itself, only about their IDEs not functioning as they were supposed. Linux users were all programmers, but none of them were Linux gurus, like at all. They could use a terminal a bit, but all would prefer WinSCP on Windows (we had a few programmers using Windows) and MC on Linux. If faced with a tty, they would likely pause and not know what to do.
Actually, wait, there was one user who had issues with Ubuntu, not sure what was going on, probably some broken binaries or something, but the shell was always freezing. Both Unity 7 and GNOME Shell (both wayland and xorg). Moved her to Fedora and she never had an issue anymore.
As for Windows users, my lord. How many broken audios after Windows updates. And I’m talking, just normal windows updates, not feature releases. Only on windows 10 (between versions 1909 and 20h2). Again, it may be my bias, but my experience was always that Windows itself does stupid stuff to break something that could have been avoidable, more often than Linux does.
Again, working as a sysadmin and doing some Windows and Linux helpdesk on the side may not be a normal representation of users. But the fact still remains that Windows had issues, either with simple stuff like installing a non-functional audio or wifi driver which required uninstalling and reinstalling the official one (Dell Update Center really saved our asses big time, because it made it really easy), or really strange stuff like plugging in a monitor or projector and not detecting it (which was because windows had “uptime” of more than 30 days, but that’s only because Windows refuses to do clean reboots if you don’t do them manually from time to time). Or there were times when Windows had 100% CPU utilization for no reason, with all updates done. At other times, indexing started misbehaving and was trashing m.2 SSDs at 100% utilization in task manager. Personally I’m sick of Windows shenanigans, which is one reason why I don’t want to be the one to have to repair it and why I try to push Linux on the users I have to manage.
You know when Windows works best? On older laptops and PCs that don’t receive any driver updates anymore, with maybe the exception of GPU driver updates. But that is also where Linux shines too, even more so because it’s lighter on resources. But nowadays any post-Sandy Bridge PC or laptop can run Windows just fine if it has 8 GB of RAM and an SSD (which is not always the case). But not for long (thanks MS for W11).
Well, speaking of W11, to be honest, if what they say is true and there won’t be anymore BSODs happening on 8th+ gen Intel iCore, and 2nd+ gen AMD Ryzen, I’d be actually happy, because less people will have issues, meaning less things for me to solve in the future (unless I decide to be a d**k and refuse to repair Windows PCs, lul).
I don’t know, changing default audio devices was always easy in KDE (Manjaro and Fedora) and on Ubuntu (Unity) for me. Not sure how GNOME Shell does it these days, as I moved to the terminal before I had a chance to try it for real (heil pulsemixer).
TBH, I still have my doubts about gaming on Linux. I’ve yet to build a new PC and start playing games on Proton and Lutris, and only used Wine and PoL like idk, 4h total in 10 years. I will do the gaming on Linux challenge, but as a Linux guy.
TBH, I think this gives the best of both worlds. But I always recommended having 2 PCs and maybe a single monitor with multiple inputs. That way, when you want to play games, you switch on your Windows machine and do your thing, and after you finish, you turn that off and just switch display input. That’s what console people are doing and you don’t hear them complaining that they can’t play their PS5 games on their PCs. I’d say to treat a Windows box like one would treat a console, and your Linux box as one would treat their main PC. GPU passthrough centralizes stuff, but I feel it’s kinda inefficient and TBH, kinda janky. LookingGlass appears to remove the jank and make for a more seamless experience, but still, I recommend 2 PCs. The Linux one doesn’t need to be super wow or something, an 2nd - 3rd gen NUC would work just fine.
Again, people shouldn’t convince people to switch, just help them if they desire to switch.
Edit: am I developing dyslexia or something? AAAGH, SO MANY EDITS!
I’m afraid the “Jailbreak your OS from a website” approach to installing Linux Distros would not fly, especially for IT security.
I mean, we already have the USB stick idea in those bogus “Fix Me Sticks”, but nobody knows some of them are actually Linux based.
Linux is already quite easy to install. That’s never been the problem.
If you want people to switch off of Windows, the main thing you have to do is switch them off the proprietary Windows-only programs they’ve become dependent upon. If all the important programs someone uses have almost-identical Linux equivalents, it’s easy to convince them to try the switch. If not, there’s likely no way to get them to change.
Serious gamers are never going to switch to Linux, but casual gamers might consider it after seeing Steam on Linux, the selection at GOG, etc.
Linux has good productivity and multimedia apps, and it helps a lot that those are freely available on Windows as well. Plenty of Windows users already use VLC and LibreOffice. Those who don’t should be introduced to them. GIMP and ffmpeg are less well known and most people would benefit from finding them. Other Linux-compatible tools may be very valuable, too, but that depends on area of interest… e.g. Accounting, 3D modeling (Blender), CAD (AutoCAD), video editing, communications (Linphone, Jitsi, etc), etc. And that’s not even getting into web browsers, which were an big issue in the distant past but no longer any concern.
Once over that hurdle, you get into hardware compatibility. Linux can be made to work on most anything, but some systems will work very poorly at a given time on Linux, while many others will work perfectly.
In both cases, advocacy could help. A website with a community around it could collect, track and provide all kinds of info that would help non-Linux users take their first steps in that direction. First with applications… something like Alternative.to that only recommends software that works on both Windows and Linux. Second with confirmed compatibility with specific models of PC/Laptops and Linux distros… Possibly even the Linux equivalent of “Registry Fixer” applications, which make it a couple clicks to download and install Wi-Fi or graphics drivers, do the needed configuration workarounds for whatever issues (rather than directing new users to manually edit configs), and more. If you want to help increase adoption of Linux on the desktop, those are the kinds of things which would actually be useful.
We need better PR, for linux and a massive amount more of available content that isn’t borderline un-catering as a lot of it can be.
Having people on the autistic spectrum explain why and how to use linux is not ideal for most scenarios concerning regular people. This is not an attack on autism, i personally prefer autists as they can be so much more less annoying than actual people in many cases. It’s just a way of saying, that such people may need some help in relaying what they know and how.
Regular folk, need some of those powers that people on the autistic spectrum naturally have and vice versa, so its a win-win for both parties involved.
You can do it, but it has compromises when trying to play high res video, desktop UI lag, etc.
I’ve got a Pi4 and PI400 and imho it is “almost” there to be properly usable without those compromises.
You could do it in a pinch now but it isn’t a great experience; and unless its a pretty great experience people will be less inclined to enjoy it.
It has the potential to be a great experience for these things with just a little bit more power. That’s when the PI will help really accelerate linux uptake imho.
More importantly, for kids, I think things like RetroPi, Pihole, etc. will be the gateway drug. The “oh that’s cool” thing that gets them playing with linux hardware, and then realising that (assuming the desktop UI performance above is better) it can actually be a competent desktop on the same cheap hardware.
Cheap hardware that works well that doesn’t run windows is the thing that will tip people over. If it runs windows well there’s no incentive to run linux on it for most people.