It peaked at 4.05% in March. The last 2 months it went just below 4% as the Unknown category increased. For June the reverse happened, so 4.04% seems to be the real current share of Linux on Desktop as desktop clients were read properly/werent spoofed.
You must log in or register to comment.
uMatrix prevents me from loading this statcounter website. :-( Can’t lookup how they measure things. In the comments people assume the stats would be counted by just looking up the user agent, which is a naive approach. I don’t think agencies dedicated to stats are doing it this simple. They have way more possibilities to track and to look what browser you are using. The stats are more accurate than you probably think. They do not need to know the exact version of browser you are using, just which type and maybe what operating system you are on.
If a script for Windows does not work on Linux and fails, then they know you are not on Windows in example.
Why are you still using uMatrix in 2024? Wasn’t it discontinued 3 years ago or smth?
It’s still useful with all its functionality. Even if it does not get updates, it still blocks most stuff and I can enable or block stuff too. Plus the blocklists it uses are still updated, as they act independently from the main addon. There is no better alternative in my opinion. I use uMatrix + uBlock Origin.
I have had to do some work on my windows pc and I hate it. I have been away from desktop for a while now and changed to linux for personal one. At work it is all G suite, which does work to its credit, but the windows OS and microsoft cloud documents suck so much. The look and feel is clunky, so clunky. Constantly refreshing and just being shit.
Never forgive forcing outlook.
I work in email Marketing and Outlook is the worst client, especially desktop. Everything I make has to have accommodations for this shitty inbox
As a service outlook (microsoft mail) sucks, as a web service their page sucks andnisnpoorly laid out and optimised and the new desktop client is atrocious.
They made me change from hotmail outlook, created a hotmail folder for under the new outlook and now I dont get notification for that address on the outlook app. If something gets marked as spam that isnt, marking it not spam sends it to the outlook inbox regardless of intended destination. If I reply to a mail it prioritises outlook despite the mail being sent to hotmail. As a result I cant log into shared files that people gave access to one but not the other.
And as a user do I hate it too. It is too many times while I edit an email and click delete that is deletes the email instead… It seems if I click a word and get the spell window does the focus always change to the list of emails… And it also force a spell correction if I click space… I didn’t pick one of the options I just want to edit the word myself!!! And if I scrolldown to remove some parts of the email thread or just want to copy a part won’t it let me if I don’t click twice… and it jumps around…damn I hate it so much. Sorry that I replied to you with all these anger. But I really felt it when I saw yours and ops comment. I hope we one day will at least think it is an ok client to work with
All the different versions as well. Outlook web is decent, desktop is terrible. The Mac version seems to be closer to web, some problems I have are fine in one version of outlook and only appear in another. Why can’t the app just be a container for the web version?
1980s: Hey guise, computers are now cheap enough that you can run an entire system and all your programs on your own machine at home instead of having to dial in to the mainframe!
2010s: No, we’re putting it all back on the servers, you get a thin client.
No technical reason btw, just because “fuck you”.
In theory I guess it provides better security in some ways, but certainly not all over giving you hardware and a VPN. So there’s that. But yeah, it sucks.
From a macro economic perspective, (and im not advocating for a conspiracy, just aggregate business interest) they’re dropping energy usage so they can pay less on their electricity bills.
So actually a double fu. get less so they can pay less rent, to provide lesser service.
Because rent seeking is the only tech bubble left.
Fwiw, my blog’s statistics say Linux is around 10% and I know a lot of browsers identify themselves as running on Windows when they’re not, so I wonder how it’s measured.
What’s is the main topic of your blog?
Travel, nothing tech related
I would wager thats your audiences bias showing up. If you did that measurement in lemmy users, you would get likely 90% Linux users.
lol no you wouldn’t. I’d be shocked if Lemmy users were greater than 20% Linux.
You can tell from the neofetch screenshots of their VMs showing weird configurations like 3 cores and 6 GB RAM with 15 minutes uptime.
Weird configs aren’t always a tell, my daily driver is a desktop with a tigerlake mobile engineering sample cpu
Are you implying that the host OS in those situations is not Linux?
Yes. Probably VirtualBox.
I think the venn diagram overlap of linux users, and users of adblockers or noscript users that block such tracking is quite large. Must impact the statistic significantly.
I use ad blocker but do not spoof user agent. If uBlpck does not block statcounter, I should be counted.
We should do a poll, who uses Linux?
Seems like you will get a biased sample here…
Why?
(Joking)
Sorry, I stopped playing factorio on my work Linux computer. I will play next month to get us back up.
Don’t worry, I’ve been playing plenty of Linux factorio for the last few weeks to make up for it.
Thank you for your service
In all seriousness, I think government bodies switching to Linux (UK’s, China’s, some Indian states’) attributes the most to this.
I’m fairly sure it’s deficiencies in StatCounter’s measurement that’s accounting for it. Statistical noise, basically.
It’s probably even higher than that. These stats are mostly based on website visits I believe. And many Linux users are also privacy-minded and might spoof their OS in the browser. I bet a large portion of the Unknown is actually Linux too.
And a portion of the “Windows” as well. Hiding in plain sight and all.
It’s hard to tell, as there are so many things that influence it. A huge factor is selection bias, as only a small number of website embed StatCounter, and that’s very likely to not be a representative sample. I’d bet that the influence of that is magnitudes larger than of user agent spoofing.
Even if that’s the case, it’s telling of Linux’ maturity.
Oh absolutely!
deleted by creator
But that’s not really a Desktop is it? If we’d count mobile device we’d also have to include Android and then the situation would look completely different.
Steam Deck is a desktop. It is exactly the same PC hardware and software you are using on your desktop PC. It runs the same games and is software compatible. Steam Deck is a desktop PC.
Android has a different hardware (not x86 compatible), is focused on phones, its eco system of software is not compatible with PC and in reverse does not run your PC software. Android based smartphones are not a PC.
So your definition for “desktop” is if it’s an x86 compatible architecture? Seems pretty random to me. Btw, there are x86 android device. IMO a desktop is something on the top of a desk to do typical “office work”. PCs, Macs, Laptops, etc. but calling a SteamDeck game console “Desktop” is pretty dishonest I think.
PC in the first place means x86 architecture. Desktop is the operating system, here a Linux desktop or Windows OS. Nothing random here. Steam Deck is a Desktop PC with form factor of a handheld. Just like any other Laptop is a Desktop PC in a notebook form factor. It runs, plays and uses the exact same hard and software as a Desktop PC. Calling Steam Deck PC not to be a Desktop is pretty dishonest. Also saying desktop PC must be on a desk for typical office work is random too.
Also this is all about stats. Don’t forget, this is not your opinion or my opinion. I am speaking from the stats perspective, as the Firefox browser (or any other browser) is just a Desktop PC from the perspective of the stats.
If you say I am pretty random and being dishonest, then first, you are discounting others opinions, secondly did not even understand the purpose of my reply. Maybe you should watch out your language next time before responding if you want further discussion. I hope my reply here made it crystal clear.
The names are pretty clear and are about form factor. Desktop is something on top of a desk. Laptop is something on top of your lap. Hand-held is something you hold in your hand.
The steam deck is a hand-held game console - doesn’t matter what OS is it uses. It’s true that most stat tracking sites count it as “desktop” but not because it’s a desktop computer but because the user agent looks similar to desktop user agents.
If I install Android on a tower PC it doesn’t randomly become a smartphone even though all browser trackers would register it as a smartphone.
And Valve using a “typical desktop OS” on their handheld console doesn’t magically turn it into a desktop PC.
The example with Android makes no sense as a argument in our discussion.
The steam deck is a hand-held game console - doesn’t matter what OS is it uses
Wrong. It does matter. Plus, it matters what software and what configuration it uses the moment it gets counted in the stats.
The discussion is not about what you try to tell. You can say what you want, it does not matter that the Steam Deck uses hardware that is PC and software that is a hybrid of Desktop and Gaming operating system. And when going to the web with a web browser, then the Steam Deck is in Desktop mode, and therefore seen as a Desktop PC. Not only act it like that, the functionality is a Desktop PC.
The Steam Deck has two modes if you forgot that, and we are talking about the Desktop PC mode. Which is what the stats are all about.
But how many use it for browsing, which I imagine this data is from?
deleted by creator
Valid point to be honest, but the answer is probably more than you think. I have a PC and still used the Steam Deck to browse the web too, not at least to install stuff. Also searching something while playing is useful too. Its made to be docked to bigger screen as well.
While you are probably right, my point was its still a PC, because he compared it to Android. And why this is hugely different. His point was to exclude Steam Deck, because it is not a PC, just like we would exclude Android. This data from the stats probably does not make a difference if its a Steam Deck or not (nor can it tell it? because browsing is the same as on PC, its an Archlinux and regular browser after all). On the other side it can definitely tell if its Android and exclude it.
So regardless if you think people browse the web with Steam Deck or not, this data should not be able to tell the difference between most distributions and Steam Deck, as its just a normal PC with Firefox (or other web browser) from the point of the stats. Just my assumption.
Connect the Steam Deck to a compatible dock and you can quite easily use it as a desktop. At the end of the day, it’s still an x64 based PC that’s just handheld.
I’m not sure that’s really a good argument. I can connect an android smartphone to a monitor, keyboard and mouse and call it Desktop. It’s also just an arm64 or x64 based PC just handheld.
A Desktop PC IMHO is a device that is used for everyday “office” work and neither android smartphones nor steamdecks are that - but laptops for example are (IMHO)
Steam deck has a full fat kde desktop on the stock os
I could install a full fat kde on the entertainment system of a car - still wouldn’t call it a desktop PC.
You can plug in a keyboard, mouse, and monitor
It is. I use it as such regularly. Keyboard+mouse+screen = browsing firefox as usual. Works quite well. Libreoffice, okular, signal desktop… I’ve used worse computers in recent years, steamdeck desktop experience is better than many 4 to 5 year old cheap laptops with win10 or win11.
deleted by creator
no, the statistics are based on browser agents, very few steam deck users browse the Internet on their devices. it’s also only half the Linux devices on steam, not of all Linux desktops
In Steam maybe. But this is StatCounter which is website visits. I doubt many Deck users are browsing the web.
deleted by creator
Half of the Linux machines on Steam, not the entirety of Linux.
A very important distinction
deleted by creator
Source? Last I checked, the Steam Deck was very much in the minority even when narrowed down to just desktop Linux.
deleted by creator
Source is the Steam hardware survey set to show Linux data only. He forgot to mention the statement is only true for Steam Gamers, not for all of Linux desktops outside of Steam.
I believe the Steam Deck would be a significant portion of the Linux desktops, but Steam’s survey might be a biased source, still
What does significant portion mean to you?
I can imagine it’d affect statistics enough to see a clear change
Oh that is a good point, why didn’t I think of that!
The youtuber matt from thelinuxcast sucks.
I am regular user, i don’t code for living and my job is not tech related. I wanted to try linux and many of you guys supported and now I’m using Linux since 2 weeks its linux mint. That matt guy was so against linux mint that i thought it was shit too. But when i installed and started using it. It has been a smooth journey. Many people in linux community were helpful. But people like matt really make it for us regular guys scared to use linux. I really hope many good linux user help regular people switch to linux and increase this number.
I get what you mean. I see a decent chunk of often more tech-proficient Linux users putting down Linux Mint, and it saddens me because even though I don’t use Mint anymore, it was still the first distro I properly daily-drove and I still consider it an amazing system for people who are new to Linux.
I’m very glad you’ve been having a good experience with Mint!
I don’t get them putting Mint down either, and I’ve built multiple Gentoo systems… I don’t need an easy distro but still use Mint and like it for what it brings (basically, it’s Green Ubuntu, what Ubuntu was supposed to be before they lost their way)
This is an interesting post. I know the YouTuber you are talking about. I do not really like him but I would have considered him harmless. It is interesting to get your take.
Mint is great and quite popular. So, if you are going to make a YouTube video saying otherwise, I guess you have to go out of your way to come up with reasons why. I am sure that makes it sound worse than it is. Something to think about.
I have distro preferences too but most of the differences really only make a difference to those already using Linux. It is a bit like arguing about Ford vs Chevy in front of people that have never seen a car ( or truck ). In the grand scheme, they are both amazing and mostly the same thing. To listen to a fan though, one is God’s gift and the other is trash. Same with Linux distros.
Mint is great and is absolutely enough for most people using computers, still as of now. It comes with its limitations though:
- By default it runs pretty old kernel. This is fine if your hardware is at least 3 years old. It allows to easily switch to newer kernel with just few clicks, but I expect newbies to not be aware of this at all. Oh, and I don’t know if it offers some custom kernels like tkg etc, which some might want to squeeze best gaming perf etc.
- Cinnamon is still limited to X11. If you have multi-screen setup, VRR, mixed refresh, mixed DPI etc, it’s better to switch to Wayland. Plus, Xorg server gets less and less maintenance and development. All the innovation moved to Wayland, so the experience on X will remain pretty stale.
- The Ubuntu base makes it so that for 3rd party software you either need deb packages or PPAs. Some will argue (me included) that it’s not the best solution
All of the above can easily be irrelevant to you and Mint is just perfect for what you need. It’s important to point out limitations of that choice, but crapping on it because you don’t like it is just pointless fuss
On your last point, there’s also Flatpak which is available right from the baked in software center… That’s not without its issues too, but they’ve been an overall smooth experience for me so far
Yes, Flatpak fixed a lot of the old shenanigans we used to have when everything was either native package, or a binary to hope for the best and install libraries manually, or source code to collect everything that’s needed for building and again, hope for the best. It is however designed to provide a way to install graphical apps, but can’t handle everything native package does (like out-of-tree kernel modules, CLI utils, system services)
I believe it can do CLI, but that’s not always been the case and not a lot of CLI apps adopted it as a result
But for most of what the typical user, or even a lot of what a technical user, needs, it does a good job
Do you have an recommendation for a distro? I wanted to use mint, but i probably need wayland for a multiscreen setup with different scalings.
ubuntu
Mint should have Waylant support if you don’t use Cinnamon; I know Xfce has Wayland support (though I don’t use it, they can pry X11 from my cold dead hands)
atm Mint only has experimental wayland support, i tried it an got instant graphical issues on the desktop. :(
Neither Cinnamon, XFCE or Mate have stable Wayland support. You need GNOME or Plasma for that if you want a desktop (or wait for the new Cosmic desktop and new PopOS)
deleted by creator
Well, if your GPU is NVIDIA, you will also need a bleeding edge rolling release distro for now. Other than that, anything that ships recent version of KDE Plasma or GNOME (the first one handles Xwayland with DPI scaling a bit better imo and is generally more functional)
I keep hearing people say this. But I’ve got an nvidia card, and I just went with the default Mint Cinnamon install and I’ve had no problems whatsoever. I guess maybe my card isn’t new enough to run into whatever problems other people are talking about.
… Actually, there is one minor annoyance. I get lots of nvidia flatpak updates; and they are large downloads. I’d prefer not to be downloading gigabytes of graphics card updates every week. But other flatpaks demand that I have the latest nvidia stuff, so … I guess that is an nvidia annoyance that I experience. I don’t expect that to be fixed by a bleeding edge distro though!
Because it always worked on X11 and Mint Cinnamon is just that. I used NVIDIA graphics on X11 in 2007 and, apart from the extra dkms driver that could break at times, it was fine, and much better anyway than ATI/AMD with proprietary fglrx driver. Rest in piss son of a bitch.
The question was what would I recommend for Wayland. Only the brand new 555 driver combined with most recent compositors (and other packages like mesa, xwayland,…) offers decent NVIDIA experience. It’s a matter of new distro releases around this fall.
Yeah, Flatpak installing user-space driver for itself is unfortunately not solvable until there’s open source driver that is part of the Mesa project. Every time you update the driver in your system, Flatpak must update its nvidia-utils too, because their versions must match exactly. For Mesa drivers, Flatpak also installs the drivers as Flatpak, but they’re compatible back and forth and it only updates when it ships new version.
The cleanup should be more automatic, but try
flatpak uninstall --unused
Cinnamon can run Wayland in experimental mode. It’s just an extra click during login. Mint also has direct support for flatpaks repositories, with flathub by default directly on the software center.
Any experience you can share on how complete and stable is that experimental session? I probably wouldn’t throw newbie on that
There’s a bug where flatpaks seemingly disappear from the system the first time you run Wayland. But it resolves with a reboot. It happens too if you change back from Wayland to X11. Other than some minor glitches from very old software that hasn’t seen an update in decades, it runs perfectly fine.
I assume the problem is hardware. Matt’s hardware didn’t work well with LM, therefore Matt thinks LM sucks… I do wish there was better hardware support but it’s the reason apple went with 1 product = 1 OS = 1 general set of hardware. Sure not every iPhone has the same hardware, but that’s why they have the model numbers, and it’s so much easier to test 200 model mixes than 2,000,000 (Android). Windows gets all the debug info sent directly to them like the others but they also have a huge stack of hardware they can use or they can buy it to test.
Did Matt try putting the regular build on a newish machine? That’s what I did with my current and was struggling until I put the latest kernel on it, should have gone with Edge, but had little trouble after)
THE YEAR OF THE LINUX DESKTOP
the year of the unknown desktop!
Now you jinxed it
People have jinxed it 20 years in a row
Question: Why is BSD so low? (And why/what is unknown?)
The BSDs got screwed over by a lawsuit in the 90s that made a lot of people hesitant to use them (coincidentally leading to the creation of the Linux kernel). Inertia carried it from there and Linux ended up getting more hardware and software support, which is the primary reason that people pick Linux over the BSDs now.
Less hardware support than Linux without enough substantially better about it to make it anyone’s clear preference. Which isn’t to say it doesn’t have advantages over Linux. Just that the average BSD user is going to be able to easily swallow their pride and run Linux if things went wrong with a BSD install (trust me, I’ve literally done this, these people do exist)
There really isn’t a compelling argument for BSD other than interest and hobby. It doesn’t have the industrial use case Linux has.
Because it’s not overly popular as a desktop os, you are far more likely to see it in certain appliances and server applications etc, none of which will show up in a pagevisits based statistic.
404: Not found
Can we commit to only posting about round number percent changes?
What’s interesting is that multiple trackers are now saying it’s above 4 percent. Last time something was posted, people questioned the data and where they got their data (which they should). Now there’s multiple sites showing a real increase.
No, we have to post these when it’s the year of the Linux desktop
No, we will not stop
4.040000000000000000000001!
We could contact that person as a community
Based on a world population of 8 billion, that would be roughly 0.000000000000008% of a person. It’s also not even representable as a 64 bit float so I had to do this math in my head (Calculator just says 0)
So you’re saying that that number keeps going up as I get closer and closer to the actual weekend when I install it as my daily driver?
OK, who did you guys bully over basic tech support questions?
why do we negate chromeos we are 10% at least so there we really are
For starters doesn’t it only run apps from the Google play store?
You can “sideload” apps outside the google store.
For example I’ve been using myinsta instead of IG. It was very easy to download and install.
and what kernel would you say it and android runs? relevant?
This is the format collision discussion that has no solution so far. A tablet that runs windows is counted as Windows. A laptop that runs android does not. Neither does an android cellphone. It all boils down to web browser user agent fuckery. This is why steam’s numbers are more reliable than other sources, they’re direct hardware surveys.
But the point is that a steam deck is not (but in a way it is basically just) a PC. There are tablets than run desktop interfaces and now there are laptops that can be used as tablet. Eventually the artificial mobile vs. PC/desktop/laptop schism will stop making sense.
true, but in this case you can look at the same graph that was linked and see another Linux distro clearly marked that they choose not to treat as one in their headline. seems a little silly to me
The combined forces of microsoft reaching new heights of greed and intrusion, plus the massive dev efforts for the best ever GNU Linux and Proton 📈📈📈
Whats interesting is that both income , profits and the stock have been growing well for years, maybe they are just monetizing more aggressively because they can’t compete on product quality (unlike other markets that are still evolving, AI and Cloud). not a ton of stuff to improve in operation systems it seems.
There are system dialogs that have unused space for ads, still plenty to improve!
In all seriousness, cloud (azure) and office subscriptions blew up and account for like 70% of MS profits. They know the Windows experience is lacking, but when they already capture so much of the market and it’s such a small slice of revenue, they have no incentive to improve.
Microsoft set themselves up the bomb
All your corporate greed are belong to us
Microsoft: As long as Adobe, Office, Autodesk are on my side, I have nothing to worry about.
dont forget soystemd and wayland
i don’t have a strong opinion on systemd, i just heard someone call it soystemd in some YouTube video once and it just stuck in my head for years
I use Linux for personal use, and Ive been using windows for work due to necessity.
There is one app I need that does not support Linux. I contacted their support asking about a Linux version and they suggested using waysroid to run the android version of the app.
So, when I have free time I’m working on switching things over.
My main motivation is Microsoft pushing ads everywhere and being aggressive about using online accounts and stuff that like.
I switched to librewolf so I will now show up as windows
Yeah me too, safety in numbers. Maybe if Linux desktop gets bigger than Windows they’ll swap it around 👨💻
Same, there must be a percent or so of Windows actually being Linux instead.
I unspoofed Librewolf back to Firefox + Linux. That way I’m not contributing to Chrome and Windows market share and perceived dominance. Plus the more people don’t spoof, the less of a need there will be to actually spoof at all as the Linux market share increases.
The problem is that websites sometimes change there behavior.
Do you have any examples of this?
After seeing Garuda Linux set my user agent to Windows, I set my Windows install user agent to Linux.
Seeing Twitch.tv login break after changing my user agent was hilarious
What did it say? Why does it break?
I wonder what’s on the ‘unknown’
likely content blockers preventing the trackers from working properly and invalid user agents. So i would expect about the same ratio of usage on there as well. Maybe very slightly more Linux since maybe the users are more likely to tinker with their browser configs and install content blockers, but even there Id say its an extremely slim minority of even linux users who do that
StatCounter also sometimes miscounts when new versions of windows or macos come out. At one point (I think at windows 11 release) there was a huge dip in windows 10 users and a huge gain in “unknown” and it was quickly fixed.
Amiga.
That 6% attributed to “unknown” is the one true OS, the only one ordained by the Almighty… Temple OS!
True Temple OS has no networking
It does have networking if you truly believe!
Heretical, you will burn in hell
Based security approach
Bare Metal, they are injecting Ethernet cable directly into their bloodstream.
In essence Netrunner then?
Case on his Ono-Sendai
Netcat, mostly
