Unfortunately, Cloudflare does not agree if you filter for "likely human", which results in 6%. If you include bots, it hits 9.3%.

https://radar.cloudflare.com/explorer?dataSet=http&groupBy=o...

If you increase the date range to 12 months, you see that (with "likely human"), linux usage jumped from 3-4% to 5%-6% in the last 1.5 months. Something changed.

ChromeOS jumped down several points in the same time period. So maybe some update of ChromeOS got misconfigured?

> ChromeOS jumped down several points in the same time period. So maybe some update of ChromeOS got misconfigured?

It's schools letting out for summer. Most Chrome OS users are schoolchildren using their school issued Chromebooks. We will see it jump back up from ~1.5% to ~5% when school starts in September. The same pattern happened to Chrome OS usage in summer 2025.

Is my understanding of that graph right? if the periodic change is the weekends, that means the total share percentage of requests from Windows and macOS users drops, while Linux increases in share percentage. Linux traffic is stable regardless of weekday/weekend cycles compared to Windows and macOS? Otherwise if there was equal drop in Linux, then the share percentage would be a smoother line. Is that right?

You can't tell from this graph alone the extent to which the change in share during weekends is due to decreased use of the more popular operating systems or due to increased use of Linux.

It’s still really strange that Linux is several times larger than ChromeOS. I would not have guessed this. And I daily drove Linux for years.

You have to buy a Chromebook to run ChromeOS, right? You can install Linux on your PC as dual boot. The investment is far smaller. However you have to buy a Mac to run MacOS so maybe that means that a Chromebook is not a very compelling machine to buy, or has a limited use case.

I don't believe this for a moment. I have seen many more Chromebook users than desktop Linux users. Not individuals, but a lot of schools use Chromebooks where I am. They are probably not well-tallied in the results, because they are often used to do some work in Google Docs/Sheets, plus some education-specific sites and less browsing on the general web.

I guess my point is that it is nearly impossible to get reliable counts from user agent statistics, etc. (lots of crawlers, that identify as a regular desktop as well etc.). The best way would probably to do real-world surveys among random population samples. But that costs money/time.

I never saw one on this side of the Atlantic Ocean (Europe) but I do see a few of them in the malls. I asked to a high school professor and she told me that she does not even know what a Chromebook is. She googled for it and "It's a kind of PC?"

If you increase the date range to 12 months you can see that ChromeOS is several times larger than Linux but recently there must have been an update so ChromeOS is now incorrectly detected as Linux.

In what universe would I jump from a proprietary OS to another proprietary OS without a damn good reason?

One where you don't like the first OS, believe the second OS does what you want, and don't care whether it's proprietary?

6%-14% (throughout the day, for some reason it's most popular late at night, Steam Box and other gamers, maybe?) is still the largest it's ever been, though, and it's presumably not getting smaller.

It's about the same as Firefox share, which is getting smaller.

6% is still significant. I'm used to seeing 3-5 in that slot. Hey, I'll take any increase, even a single percentage!

I imagine a big part of the puzzle is that many people do not even use laptops/desktops regularly anymore, just phones and tablets

I guess almost all laptops/desktops are corporate, student, gaming or own business use.

I have never, ever, seen a corporate linux desktop outside of devs.

I have never, ever, seen a student using a linux desktop apart from CS students and a few physics/maths.

Linux is not for gaming.

See above answers for Linux in own business.

Just my anecdata, I've seen dozens of CS students that own a Steam Deck in the past few years. It's a perfectly suitable alternative to owning a gaming laptop for many students.

That, and the endless stream of automated requests from various crawlers identifying themselves as browsers.

Also, Chromebooks.

I don't think I've ever seen a single chromebook in the wild being used by a real person for real. I am sure they exist of course. I must have even seen one and forgotten it. But in all the years since they have existed, and all the varied people I've come in contact with in all those years, if I can't think of any, then the percentage is low.

I think a relative's kid had one issued in high school a few years ago. There you go. One out of who knows how many hundreds, and that one was not even a voluntary purchase. In fact that same kid built himself a budget desktop gaming & streaming rig when deciding what to do with his own money. So I wouldn't even call him a chromebook user. The chromebook was just school equipment he had to operate like a copier or projector.

Yea Chromebooks are wildly popular in American education. They've replaced the school desktop PC.

My wife has one. Been rock solid for years. I recommend them to relatives who just do email and browsing etc.

I used one in the wild, but I was working for Google at the time.

It would be hell being forced to use these things for your job. What departments were? Because I can't imagine a software developer using one.

I wasn't forced to use anything. This was in the mid-2010s when the first ARM Chromebook came out; I bought one out of curiosity. But I learned about it at work.

Independent of that: some of the modern higher end Chromebooks can be quite nice, and you can (A) run general Linux on them, and (B) VS Code dev tunnel in the browser runs on them just as well as on any other laptop.

yeah i forgot about their linux capabilities they got. In my mind they didn't seem a lot different than simple android tablets with a keyboard but indeed if the hardware isn't bad it could be valid option now I think about it.

> In my mind they didn't seem a lot different than simple android tablets with a keyboard [...]

Well, that's also what a MacBook Air is. And a MacBook Neo is the same but with a phone's processor, instead of a tablet.

But that first tiny ARM Chromebook I tried in mid 2010s actually sucked a lot. It was cheap, though. It was the time of the netbook craze.

Yeah but these MacBooks run MacOS, which is a fully fledged desktop OS compared to ChromeOS which looks and feels a lot more like Android and lacks features in some areas.

ChromeOS is a Linux, and you can run anything Linux in them, including all the command line apps you like.

Beyond what was provided natively in ChromeOS, you could often get something like GalliumOS working on them. My old Chromebook was fairly painful to use as a Chromebook, but GalliumOS made it pretty snappy and fun. Unfortunately that project seems unmaintained now

Vscode can run in a browser, your code is often in git, and it normally is going through CI/CD (being compiled/packaged/deployed elsewhere).

All you need for that is a dumb terminal.

Boy I hope VSCode runs better in the browser than all of Microsoft's other terrible web apps.

I have seen several because family members have owned them. They are nondescript black rectangles that are owned by cheap people and kids. There is no reason for you to notice them in particular and it's useful to remember our perception isn't a representative sample of anything.

The problem is, we have no way of knowing if our personal experiences are actually representative of everyone else's experience. I have 2 Chromebooks but I'm a computer nerd so I know I'm not your average user, but going off 2025 shipment data, it's something like 223 million Windows laptops to 25 million Mac's and 22 million Chromebooks.

The separation of “OS X” and “macOS” gives me pause about the reliability of the other figures, particularly Linux and Chrome OS.

The last release of OS X was 11 years ago but this site shows it having over twice as many users as macOS. It's totally implausible.

I haven't done any investigation but some scraper and/or bot configuration uses OS X as its user agent.

I run an unpublished service used by 1s of people and was hit by an influx of supposed OS X user agents that didn't correspond to my actual user base.

Did you run any advanced TCP fingerprinting techniques to dig a bit deeper?

Well, guess it's worth spending a little time looking at it.

It appears that it's the Cloudflare Security Center scanning service? This would also explain how my unpublished service was located as I use Cloudflare for my DNS.

Regardless, they're the same OS, so any serious (read: competent) person compiling stats like this would combine them into the same bucket.

I saw reddit and turned around already

I guess Steam OS counts as Linux?

That's because it is Linux, specifically Arch.

All Linux is Linux.

Except for Android and ChromeOS. Nix is kinda weird too. But other than that, yeah.

If it uses the Linux kernel then it is Linux, that's where the name comes from. Android is no less Linux than RedHat or Debian is (when Debian is using Linux rather than FreeBSD kernel or Hurd, of course).

I see people didn't get my point. If StatCounter considers Chrome OS different from Linux then they might also consider Steam OS a separate bucket. Or they might include Steam OS in their Linux count and thus the rise of Steam OS might be causing the increase in Linux share.

I don't know if the BSD folks would be happy to see it mentioned or unhappy that it's at 0%. At least you know that most of those users are human

Netcraft confirms it...

Source: vibes

Imagine first thinking that humans are using Linux on the Desktop after seeing this (without specifying which distro) and then realizing that it is bots which are the "desktop users" because they are using Chrome to scrape websites since forever.

Any reason to think that bots are linux and not windows/macos. If you look at crawling resellers, you can simulate mac/windows/ios/android. Why would crawlers prefer linux over those?

Only 1/3 of those are not "human".

Depending on how this data is being collected, a factor could be the rise of search / fetch / scrape capabilities, added in lots of places. Agentic this, google killer that, and so on. The easiest way to parse web content nowadays is to load up a browser and drive that. And thats easier to do with linux / mac distros, both at scale (i.e. datacenter based solutions) and at home.

As a bit of a side note, this year I started noticing captchas / cf verify speed bumps while browsing from the same residential ISP with the same mac laptop. I never used to get them. Maybe has to do with all that "buy a mac mini" craze that happened a while ago.

Yes, the dam has been broken. Major linux distibutions are getting polished for many decades already and they are totally ready for the mainstream adoption. No amount of marketing will be able to reverse this trend. Considering that Big Tech is spending all their earned money (and even more by borrowing from investors) on AI, nobody will be even trying.

It's obvious that Microsoft doesn't even want to be in the consumer desktop OS market anymore, they'd rather pivot to being pure B2B like IBM did. Likewise, it's obvious that the future of MacOS is continued iOS-ification and infantilization of their OS until it will eventually be useless for anything except developing iOS apps. So neither major player will bother spending a marketing budget to reverse this trend; they'd rather not play the game at all. On the flipside, fewer and fewer people will have traditional desktops in the first place; owning only a phone and no PC at all is the new normal.

What have they iOSed about macOS? The latest versions haven’t lost any features as far as I’m aware. But Windows lost the ability to even pin the task bar to the side of the window in Windows 11… and the ability to stop auto updates.

If anything, iOS is getting more Mac-like, especially the iPad version. It has multi windowing and a window manager now.

Everything. The whole OS has been getting locked down more with each release. More harassment about Apple accounts, harder to run non signed apps, the OS gets less and less configurable. I mean when Launchpad was released, it was clearly and iOS style interface to find your apps. Maybe you aren't old enough to remember when you could literally do anything you wanted with the OS and make it work for you.

> What have they iOSed about macOS?

How about this? https://news.ycombinator.com/item?id=25074959

Meanwhile, here I am posting this using Google Chrome, with Lightroom Classic, InDesign, Sublime Text, Steam bla bla all opened in the background.

You posted a 6 years old link.

The post is still relevant and it's not about not being able to open non-Apple software, but about Apple contacting their servers before anything is started.

Apple is not blocking a lot, but step by step they're building their way into a locked OS, and it's very obvious.

> it's obvious that the future of MacOS is continued iOS-ification

What’s an example of a specific iOS-indication in macOS that makes the OS worse?

Launchpad. It sucks on macOS. You can just not use it but it's there.

Fullscreen apps and Spaces. My Mac isn't an iPad, but their window manager wants me to use it like one.

Hidden scrollbars.

App sandboxing.

I think people don't know what OSX was like, when it was a glorious Unix system that did what you wanted. It let you install extensions and run unsigned code. Now its just a boring locked down iOS that has a windowing interface.

> Hidden scrollbars.

System Settings | Appearance | Show Scroll bars | Always

OS X: 21.14%

macOS: 8.6%

Yeah, if you believe this kind of stats it's on yourself.

Also why are you posting a random reddit link? It doesn't even match up the latest stats provided on the official source: https://gs.statcounter.com/os-market-share/desktop/worldwide...

It does match statcounter if you filter for North America.

Aren't macOS and OS X the same thing?

> Launch: Released in March 2001 as Mac OS X 10.0 "Cheetah", replacing the classic > Mac OS.Name Change: Shortened to OS X starting with version 10.8 / 10.9 in 2012, and rebranded to macOS in 2016.

I think that's the GP's point. If the people who put this together are too lazy/uninformed/incompetent to combine those two categories into one (because yes, they are the same thing), why would I believe anything about their collection methodology?

It’s hard to trust StatCounter, or anyone presenting web traffic stats at this point, after they showed a large spike in Windows 7 users for a few months, before it suddenly disappeared. (Even better: the blogs reporting on it with a made-up conclusion that users were revolting against Windows 11.) Bots want to act like humans, they run client-side JS, and they use residential proxies.

https://www.windowslatest.com/2025/10/02/no-windows-7-isnt-g...

Even if you assume OS X is 1999-2016 and macOS is 2016- then the numbers are still odd.

This is what Wikipedia has recorded.

I presume Chrome on Mac still reports OS X as its OS, even when running a newer version. It is such a ridiculous idea not to put that together.

My understanding is that apple officially changed from OS X to macOS in 2020 with the launch of macOS 11, even though the branding changed a few years earlier.

Sort of. Apple half-heartedly changed their version numbers after 10.15, but there were times where they wrote both the 10.X and N+1.x versions elsewhere, so I think it's just generally a confusing mess with them now.

iPadOS masks itself as macOS/OS X (depending on where you look), so that doesn't help either.

We should really be able to downvote submissions... This analysis is clearly factually inaccurate. I am curious to know why OS X and macOS are showing up separately though

The thing is, there is an ongoing linux boom (the percentage doubled which means linux user base has effectively doubled too). But god help us in this AI-slop phase.

I think the main split between hardcore "GNU/Linux" guys and the general tech world is about principle vs pragmatism.

The hardcore crowd see free software as a right that needs to be fought for, whereas others see open source as nothing more than a way to modify or improve software without requiring explicit support from the publishers.

And there are many, many parts of Linux that make the software "freer" but worse.

I remember when I first started reading HN I was surprised how many people refused to use Linux. I get it has friction but to me it was worth it for the social benefit of having a FOSS OS. That’s when I realized FOSS is kind of a lost cause if most tech people don’t even care.

People like to complain about the walled gardens and surveillance of Windows/MacOS but refuse to actually support the organizations working to make alternatives. So frankly, we deserve what we get at this point.

And increasingly, reposting crap from Reddit.

I recently googled the phrase "German Jews who voted for NSDAP" after seeing it mentioned in passing

The AI overview contained the construction "Consensus on Reddit is..."

What a time to be alive, I guess

For some reason the AI overview model likes to pick up stuff from Reddit a lot. I guess, once you get to more niche topics, it's easier to find some stuff on Reddit because Google has stopped properly indexing/ranking the small web.

Very often, AI overview states something as a fact (Consensus on Reddit would be a lot better, because it at least indicates that it is bullshitting using Reddit), but when you click on the sources, it's extracted from what a random Redditor says without any proof whatsoever.

Reddit has a deal with Google since a few years back, and they are looking to get out of it.

https://finance.yahoo.com/markets/stocks/articles/reddit-rdd...

Nevermind the AI crap, your search query requiring a disclaimer (“after seeing it mentioned in passing”) is more worrying than anything else.

What is the point of a sock puppet if you can't be explicit about what specifically you found worrying there? And I thought I was neurotic

Anyway recently I watched a video in which Alan Derschowitz compared Jews who voted for Mamdani to Jews who voted for the NSDAP. The ridiculousness of the comment aside, I was curious to what extent this was actually a thing. It did happen but it was a tiny number of fringe German nationalist Jews, <1% of voting age Jews -- but still a fair bit more than what one might expect given that the NDSAP was not mincing words when describing their intentions with Jewish people.

Those who supported them regardless included Max Naumann, who went on to collaborate with the Nazi government as leader of the VnJ movement, through which he pushed for Jewish pacification and assimilation as an answer to antisemitism. There were others -- though vanishingly few -- who collaborated in various ways, with enthusiasm seemingly beyond what was coerced.

Those who were publicly Jewish usually ended up being targeted by the Nazi regime in the end regardless -- VnJ was outlawed and Naumann was picked up by the Gestapo a few years before he died of cancer. The Nazis viewed Jews as politically irredeemable so extensive collaboration was not generally sufficient to save anyone

I do not believe this. Unless it’s some sort of tricksy word game or counting android or something.

There is 0% chance that 10% of desktops/laptops are running desktop Linux.

Login-walled for me. Redlib link: https://safereddit.com/r/linux/comments/1vcpk8i/linux_deskto...

The same people who believe these numbers also think open routers token measurements are indicative of total token demand - lol.

I don't think these numbers are correct, but desktop Linux is no doubt having its year now. So many big YouTube content creators are covering Linux content and switching to Linux specifically, since the major obstacle to switching (gaming) has now been solved. Add to that the push effect from the constant enshittification of Windows.

I think Linux is getting a lot more popular, but not this popular.

StatCounter is a joke.

Cloudflare is showing 9.2%. Steam is showing 7%. US gov is showing similar stats.

I'd pick major Linux Desktop Environments (KDE or GNOME for instance) over macOS and Windows any day of the week. I feel 3-4x more productive on Linux.

With LLMs, I've been creating little desktop apps to fill the odd gaps. For example, yesterday, I made a little offline dictation app for KDE.

Flagging since statcounter's data is just so unreliable that discussing it is like discussing last week's lottery numbers.

(Just a couple of weeks ago there was an equally implausible frontpage post showing the Firefox had more than doubled in use in the US, to 12%, but only in the US, not in any other country. My money is on scraper bots identifying as Linux-Firefox, run from US IPs to minimize the impact of geo-blocking. But when Statcounter doesn't explain their methodology and doesn't care about the quality of their data, we'll never know.)

Better quality data comes from Steam with the qualifier that it only applies to Steam users.

https://www.gamingonlinux.com/steam-tracker/

It's also showing Linux usage exploding in English speaking regions.

Does anyone has list of North America countries according Statcounter? Can't find their definition of NA whether it includes Mexico or not.