I switchee to Arch the other month, its been alright except for CUDA getting an update before the NVIDIA driver so I couldn’t run my assignment locally. But I assume that’s my fault because Arch maintainers are all care and no responsibility.
Just restore a snapshot. Or just check which packages are gonna get updated. OR just don’t update right before you have to do critical work.
If none of those work for you, then Arch isn’t for you. That’s fine too. I also sometimes get intrusive thoughts telling me to just go back to Mint. 😁
I’ve found Garuda pretty much gets you all the perks of Arch without the drawbacks and installs just as quickly as debian if not faster. And I love ancient Linux memes as much as anybody but neither Debian or fedora is much to write home about nowadays IMHO.
Endeavour is a great alternative IMHO, but Garuda’s development is definitely more skewed towards gaming and comes with a lot preinstalled/preconfigured.
I’d say it’s even more simple. Comes with stuff like snapper and zram preconfigured and a bunch of tools to do various things. I use their KDE lite version since I do not like their theme AT ALL.
For a while I would have agreed, and I used sway for years. But recently I switched back to i3 (i3-rounded) due to display issues with my AMD GPU. I started doing most of my development in the TTY, and found that switching from TTY to Wayland takes half a second and can sometimes break my GPU (until I switch between TTY and display a few times). With X11 it’s instant and without issue ¯_(ツ)_/¯. Hoping that gets fixed down the road, or that it’s specific to my GPU.
Most of those are perfectly ready for every day use without issues today. All are alternatives that bring new features and specific use cases, solving new problems, or solving old problems in innovative ways. Wayland is an active replacement to an existing technology, as the old X is expected to just not exist anymore at some point in the future. BTRFS isn’t intended to replace Ext4 wholesale, Flatpaks doesn’t intend to replace apt/pacman/etc., Pipewire does the same that Pulse and Jack but Pulse and Jack won’t stop existing. Adwaita existing doesn’t mean that you can’t use QT or GTK in your projects. That’s the difference.
As a result Wayland has the burden to actually fulfill and comply with all the features and use cases that X11 already does, with all the new security improvements on top. That’s a tall order, and until it can do so, it will be undercooked and under adopted, because they set themselves up to that bar, nobody but them is responsible for this. Is the ancient “let’s rewrite from scratch” trap that all dev teams fall on at least once in their lives. It isn’t impossible, but it always takes way longer than the optimist project managers anticipate.
Feature parity with X has never been the goal. Because most of X’s features are a legacy of the 80’ and dreadfully obsolete anyway.
I’m all for maintaining compatibility where it makes sense, but carrying over a 40 years old feature set just in case is the best way to prevent anything from moving forward.
Wayland can already do or is actively being developed for stuff that is relevant to modern systems: multi-monitor with different refresh rates and scaling, HDR etc. Stuff that X would never dream of.
Feature parity, maybe not, but use cases, definitely is the goal.
I’m just saying that if users have to run X compatibility portals to get basic functionality for every day tasks, then something is not fully baked yet. There’s nothing wrong with that. But apparently pointing it out is some sort of herecy.
I don’t think it’s heresy, but I always find it funny that an extremely vocal community shits on systemd for being a bloated tentacular monster shat should be abandoned, but praise X for being a bloated tentacular monster.
In a way, Wayland is much closer to the Unix Philosophy than X. It’s a display protocol, nothing more. Everything else should be implemented by the applications using this protocol. X has grown over the decades to include way too many features and edge cases.
Translation layers like XWayland are important and extremely useful for the transition period, but shouldn’t be taken as a sign that Wayland is not ready for prime time. If 10% the people shitting on Wayland had instead worked on adding Wayland functionality to their favorite apps (that includes you fuckers at nVidia), the transition would have ended years ago.
Counter-counterpoint: Wayland is perfectly fine and production ready and has been for several years now, as long as you’re on AMD or Intel GPUs. The nVidia drivers are still undercooked and not ready for proper daily use.
Mostly all. At work we have to use teamviewer. Remote from Wayland to others work but you can’t connect from another client to a wayland client. Tried hoptodesk, ruskdesk etc. always the same.
I don’t have personal experience with nvidia graphics. How does proprietary work now? I have heard it’s gotten great this last year? Or is it horrible still?
I haven’t found any issues except sometimes when I switch to another window out of baldur’s gate 3 and switch back again, baldur’s gate 3 freezes. Not sure if it’s the game not being Linux native or the driver.
No offense, but your argument is exactly like “electric cars are still undercooked and not ready for proper daily use because I still have to put gasoline in mine and can’t afford one”.
Sure, but at the end of the day, for better or for worse, there are going to be tons of people who simply don’t care about whose fault it is - they’re going to want their system to work.
I was lucky enough that I was finally able to make enough money to swap out my 2080 with a 6700 XT this week (and wow what a significant difference in how the Linux desktop works with AMD cards), but I have plenty of friends who do have Nvidia cards and if they asked me whether they should give Linux a try I’d have to warn them that they’re going to get a subpar experience due to it - and all they’re going to hear despite me saying that it’s Nvidia’s fault is that Linux isn’t good enough.
So when it comes to Wayland + Nvidia, hopefully Nvidia gets with the program, but otherwise we’re (the Linux community) going to be at a crossroads of whether we want to get more adoption on Linux - Nvidia is not a small market by any means.
I don’t go and try to proselytize people into coming over to Linux, but there are absolutely plenty of people who do and the mindset of “It’s not Linux’s fault, its X (ha)” isn’t exactly going to work there.
Does multi-monitor sets work yet? Does it still randomly crashes when logging out? Does it have support for touch monitors already? Is Pipewire support ready? Is the Compose key still broken? Does it handle internationalization better now? Does accessibility software like on screen keyboards and screen readers already work on it?
I love Wayland, BTW, the more secure ecosystem is a net positive. But we can’t pretend it isn’t a lot of effort for something that has no tangible difference or immediate advantage for the end user, is extra work for developers and currently has a higher potential for errors, malfunctions and missing features that are taken for granted. Again, it’s a worthy endeavor to improve something that already works, but that also means there’s no rush. We can afford to wait.
It hasn’t done that for the 1.5 years that I have been using it for.
Is Pipewire support ready?
Yes. It’s so ready that even ubuntu uses it with wayland by default.
Does it have support for touch monitors already
Yes. It, in fact, has better support than x org.
Does it handle internationalization better now?
I don’t know about the problem with i18n but I don’t think this will affect most users.
Onscreen keyboard is still a pain to run but maliit works on kde+gnome/wayland. When was the last time you used wayland dude? I am not trying to sound this argumentative. If I do, my apologies but I have been listening to these same points being regurgitated over and over again when they have been fixed long ago.
Wayland integration with most DEs is absolutely incomplete regardless of Nvidia support. Wayland causes a ton of bugs every time I try to use it with KDE. There are still bugs even with GNOME like wine applications not working or screen sharing not working. So no I will not be using Wayland until it’s ready for everyday usage, which it isn’t right now.
He’s a thought. Stop being a power nerd, stfu and let people use what they want.
My desktop crashed three times so far after updating gnome, linux kernel and nvidia driver two days ago. Not sure who’s the culprit, but I’ll blame nvidia by default.
No, unless your use case is very specific (like being an artist needing color calibration/the software you use needs to position a multi-window setup etc. And color calibration is being actively worked on should have basic support in Plasma 6 according to Nate Graham) wayland is pretty much ready for daily use. It does have annoyances but they are getting actively fixed unlike X which is barely maintained and has glaring security issues. Fedora KDE has even decided to completely remove the X server on its 40th release.
You do know that the people who make Wayland are the exact same people who made and maintained X, right? Like, they are intentionally abandoning X in order to make Wayland, and eventually X will just be actually XWayland as compatibility to transition to only Wayland.
“Unlike X” doesn’t support your argument. If X11 is barely mantained, is on purpose. X11 and Wayland are not in competition, one is the rewrite of the former. They literally have no rush to push Wayland to main stage until it can do all that X11 does, including the annoying edge use cases. Because if X11 does it and Wayland doesn’t, then people would just continue to use X11. No brainer. They need more time, that’s fine, we can all do with being a bit more nicer and gentler. There’s no rush to push adoption
There is a rush because Red Hat isn’t interested in maintaining wayland anymore. Neither red hat nor Kde/gnome are interested in supporting x org in the long run. For wayland to get better and do the things it currently lacks at it needs a greater user base and that’s why there is a rush by major people in the linux community (kde and fedora for example). Right now its at that there are somethings that wayland can’t do that x org can and somethings that x org can do but wayland can’t. Since wayland is being developed actively and is the future it’s the obvious choice and x org has far more annoying use cases that are just not gonna get fixed “unlike wayland”. Majority of the users shouldn’t have any problems switching to wayland.
The Sway implementation (not Wayland as some DEs seem to run really smoothly) sadly is still completely hit or miss depending on your exact hardware setup. I have two device (both even with nvidia grphics sigh) and one of them is just a buggy and flickering mess.
I know they officially don't. And I didn't try to say that Sway was bad in any way or that it is their fault. I was just stating facts about state of it with NVIDIA graphics (that kept me -as a long-term i3 user- from switching to Wayland).
I disagree. Sway is extremely high quality software. Nvidia is a known terrible player with FLOSS software. I hope they will continue their path of recent improvements.
No one stops you from moving to Wayland. It has it’s uses. But the FUD you’re spreading is stupid and boring. X is fine, it’s exactly what a lot of people need and it doesn’t make sense to move their DEs to Wayland only because it’s ‘new’. The fact that it took Wayland 10 years to reach any sort of usability shows just how little does it offer to an average user.
X’ architecture is insecure. There’s no isolation between windows, and each process can spy on your input. That’s just one example.
Wayland is necessary.
I’ll never forget the first time I successfully installed arch and got my I3 set up juuust like I wanted it. It felt like I did something. It was great. Fuck you!
I’ll have you know that I eat a vegetarian not vegan diet and I really don’t have a man bun (got no hair for that) … The stickers on the laptop however really felt like you took a photo of my machine.
Self hosting, at least for lemmy, is absolute trash. I have been told a few times when asking questions that, “it is expected that you are thoroughly experienced with Linux” to be able to follow the mediocre guides. And they are trash if you are a newbie.
So people like me, who would love to use Lemmy for non Linux things, am posting almost entirely about Linux problems.
What does google and bing making search worse have to do with stackoverflow and serverfaults linux users being snobs?
Or 70% of lemmy circle jerking over how much superior linux is.
I think they just meant if you’re a total Linux newbie, one who might not even be familiar with stackoverflow, trying to troubleshoot Linux via a typical search engine (a very normal thing a normal newbie who doesn’t know better would do) won’t lead to helpful results do the current enshitification of Google and Bing. I’m pretty sure that’s all they meant bro
I wouldn’t expect running a publicly accessible server on the internet to be easy or a great idea for someone not familiar with the OS they’re using. Great way to learn, though.
Yeah I’ve got a proxmox cluster and I’ve been using Linux for decades but I wouldn’t dare host something that a LOT of users are going to access. I don’t know nearly enough about netsec and I can guarantee my vlan practices probably aren’t perfect, etc.
I have a solution for you. Just don’t have your Internet facing things on the same /24 as your home stuff. Why vlan if you can just separate them by network and switch.
Or: just do it anyways. I learned most shit after everything broke. Not before.
And how do you learn, then, from this project, if people are shitty about your questions?
I’m a sw engineer. I’ve been doing every kind of application management, development, and systems design for 25 years, nearly all of that in Linux, and I still need things answered about running apps in proxmox. I’m not coming to a Lemmy community for those answers, I’ll tell you that.
I haven’t tried to get community tech support on Lemmy, so I wouldn’t know what it’s like firsthand. If people are really that difficult, sure, that sucks. But it sounds like the person asking needs to work on more fundamental linux skills than something specific to running a Lemmy instance, and the internet is full of information about that.
I get that, but here we are. It’s something I want to do. I’ve been at it for 6 months and I’ve managed to get the site working twice, but am still struggling with SMTP. Digital ocean blocks smtp and send grid breaks the site.
SMTP in general is a pain to configure. I ran my own mail servers for a while and finally gave up and used a 3rd party service. Too many problems with antispam restrictions, and things like I’d finally get it configured, upgrade postfix or whatever and then it would all be screwed again.
Not really. There’s tutorials for everything and most of them still work 20 years later.
vs you installed ubuntu 20 and now youre trying to follow a 16 tutorial.
Would recommend using Docker (container) and Caddy (reverse proxy) to self-host as a newbie, streamlines everything and only basic Linux knowledge required (although you do have to learn Docker commands).
Congratulations, you’ve discovered the struggles of learning an esoteric hobby. Often the learning curve is steep like that. And often you will encounter elitist twits trying to push you back down the curve. But they cannot keep you from knowledge. It sounds like you’re already discovering some of the rewards.
Oh yes, I’ve been using the ansible method of deploying and I have it very close, I just can’t get SMTP working. I’ve set up an account with send grid but letsencrypt keeps telling me I’ve passed the limit for certs and every time I try to deploy it says I have to wait another week to try. I would remove certs but since I’ve already wiped those out, I don’t know what they are or how to find them.
linuxmemes
Active
This magazine is from a federated server and may be incomplete. Browse more on the original instance.