Pingdom Home

US + international: +1-212-796-6890

SE + international: +46-21-480-0920

Business hours 3 am-11:30 am EST (Mon-Fri).

Do you know if your website is up right now? We do! LEARN MORE

The consistent failure of Linux to grab even 1% of the desktop OS market

TuxLinux has been around for almost two decades now. It has become a resounding success as a server OS (for example as the L in the famous LAMP stack), and more recently as a mobile OS (Android). But what about on the desktop?

Linux enthusiasts have been predicting the rise of Linux as a desktop OS for the better part of a decade. To name just one of many examples, in 2003, Siemens Business Systems predicted that Linux would have captured 20% of the desktop market by 2008.

Well, it’s now 2010, and desktop Linux isn’t even close to 20%. Or one tenth of that.

Linux desktop OS market share

The sad truth is, Linux on the desktop has so far failed to really take off. Linux hasn’t even been able to claim a single percent of the desktop OS market.

Here is the current desktop OS market share distribution, courtesy of StatCounter:

Desktop OS market share

As of July 2010 Linux had less than one percent of the desktop OS market. Perhaps this is a temporary low?

Unfortunately not. Here below you can see the last two years of Linux’s desktop OS market share. Things haven’t been moving much.

Linux desktop OS market share

We should point out that there will be slight variations in the estimated market share depending on the source and the method used for acquiring the data. Other sources than StatCounter will show slightly different market share data, but it’s always around one percent. The important point here is that the market share is tiny, and isn’t really going anywhere.

What about Ubuntu?

Ubuntu’s rise to fame since its launch in 2004 has been remarkable. It has become the largest Linux desktop OS distribution by far, but the question is, why hasn’t it been even more successful?

Even with the relative success of Ubuntu, it looks like it may have mostly cannibalized the market share of other Linux distributions instead of bringing new people to Linux.

What can be done to grow Linux on the desktop?

Here are some things that we think could be done to make Linux more popular on the desktop. This would demand the support of a more or less unified Linux community, so it might be unfeasible, but one can dream, right?

  • Center around one Linux desktop distribution.
  • Center around one GUI.
  • Polish the product (yes, you have to think of it as a product) until it shines, and make sure there are plenty of polished apps to go with it.
  • Be fanatical about consistency and usability.
  • And this may be the hardest one of all: Don’t build for Linux’s current hardcore audience of tech geeks. If the goal is to bring in new users, that approach is a big no-no since it will alienate the vast majority of the population.

Then of course there’s the marketing aspect of it, branding, building mind share, and getting manufacturers to include Linux as a viable option on their computers.

Final words

Some of the charm of Linux is that it is so flexible, and that you have so many choices available to you, but this can also be a curse when it comes to mainstream adoption. Although Linux on the desktop is a far better experience these days than it used to be, it hasn’t been able to catch up with the other OSs out there that provide a more unified front to the end user.

Linux is not in any danger of going away, so it’s quite possible that all of this doesn’t matter. Plenty of users are huge fans of the OS. Several people here at Pingdom use and are quite happy with Linux (primarily Ubuntu). Perhaps Linux will just continue to be a small niche desktop OS. There’s nothing wrong with that, not everything has to go mainstream.

(And in the meantime, people will continue to predict the rise of Linux on the desktop.)

About the stats: The OS market share data from StatCounter is based on visitor stats from more than three million websites.



77 Comments

Very good analysis. The bullet-points on why Linux failed to gain a significant traction is dead on. I think it should also be pointed out that OS X’s success might be another reason why Linux is failing to grow on the desktop market. Personally I run OS X on the desktop as it allows me to do most things I can on Linux. However, in contrast to Linux, it is polished and focused on usability (and consistency across apps).

On the servers I wouldn’t even consider running OS X though, as usability and polish isn’t that big of a deal there (Bash is polished enough =).

Push a highly-polished, extremely simple to use distro as an alternative to Windows/OS X for those users who just need a web browser / Google Docs / etc.

Ubuntu is perhaps closest to this ideal, but not there yet.

For the past month I’ve been in Linux Mint (Ubuntu + things that should be in Ubuntu like codecs, etc.). I’m dual booting w/Windows, but have only been in Windows for my digital camera pictures routine that I haven’t had time to work out in Linux. So far, I like it, but I’m running it on fast hardware (quad core, dual video cards, etc.), and I’m ok with having to figure things out. I think for most users who live in their browser for web based email and surfing, it’s fine. The problem I have switching people I support is that I’m not yet good enough with it to support them when they do have problems.

I don’t see that percentage growing anytime soon though. I think until something major happens, it’ll remain an OS for the techie crowd.

Hey, we are heading towards it. Migrating 500 Clients to Fedora :)

Greets
Marcus

I think its the curse of “choice” the article speaks of. Too many choices is daunting for anyone.. The grand unification of linux would require.. ONE DESKTOP ENVIORNMENT, one KILLER app(polished within an inch to perfection as close as any piece of software can be) for each task. Make it as simple to install and use (with all the pros of linux stability, security) and make it easy to get help, no more “RTFM, READ the WiKi, Go back to windoze, Learn to google, etc ” attitudes against new users. A very informed yet easily clickable install routine.. “where do you want this installed, what file system, what will the computer be primarily used for, would you like to optimize the file system for your hardware that’s been detected? would you like a swap file, (since you have 3+gigs ram), are you dual booting? and lots of other hand holding questions.. to make the best linux out of the box (after install) as possible.. Some distros do alot of the above mentioned things, but not all.. Then if all these things are done.. and there arent so many choices.. or.. perhaps during install, you get the top 3 choices for many many tasks.. with descriptions, like a “build your own distro” on demand for newbies.. until all this and many things i’ve proberly NOT thought about are DONE , it will NEVER be the YEAR of the Linux Desktop.. not even the MONTH… alot of people on various distro forums I’ve belonged to over the years worry about “dumming down” linux.. i remind them when i can, dont worry, while new users want something easy, as they mature (the ones that wont just WONT, just like windows users, they just want things to work, even in some cases it takes a little bit of work to make things work, key word “little bit of work”) they’ll branch off into less “hand holding” distros, then most of the mainstream hand holding and do it mostly or all yourself distros will keep their userbase, heck, even make it GROW.. until the egos, attitudes of many distros userbases changes.. and a unified linux distro comes along.. Choice will be linuxs greatest strength, and even GREATER weakness.. thats my .02 now I’m broke

There’s a big flaw in those numbers: most Linux users I know always have NoScript on and running in their browsers. That way they (we) don’t count.

V.g. myself, I have always blocked Statcounter (and other counter/profile/ads services as well) in my daily browsing.

That’s a long article to say
1. Linux has 1% of the market
2. Linux should do … blah blah blah (which the suggestions are nothing that hasn’t been said for years and has been shot down for the same number of years for other benefits).

As a Linux user, who says I want Linux to be widely adopted? Let the general masses use what they want, and if that includes anti-virus, anti-spyware, etc. then so what?

If they want to move beyond the basic Windows or OS X then Linux is ready for them.

Maybe they should have to come to Linux, not the other way around.

Hmmm.

I dunno. The article bases itself on statistics, and the statistics are a poor, imprecise portrayal of what the market share could be/is.

Even MS themselves estimate about an 8% desktop share.

The statistics do not, for example:

1. Take users of any kind that bought a machine that comes with Windoze on it but installed Linux.
2. Cover the entire sample spectrum of the internet (not all websites deal with their statistic backend- there’s actually several… And the Marketshare on one of them for Linux is actually 4.35%)
3. Deal with companies such as System76 or ZaReason- which are starting to gain a much larger user base.

Linux on the desktop is not dead yet- and I’m of the belief that it has never been.

I installed a Linux distro for a friend of mine, he loves it. A customer of mine the other day is using it (and he used Windows XP for years on end). I switched over to Linux completely and no longer use Windows myself.

I don’t mean to be a blind zealot, because there’s already enough of that- but I would certainly not trust these numbers.

About an year back when I needed to setup a PC for my parents to be in touch with us over internet, I tried to setup Ubuntu so that I don’t have to worry about anti-virus and reduce maintenance, but I gave up the idea after a lot of effort. All that I needed for them was a good browsing experience and full support for yahoo, skype and google chat. Browser and Skype were not an issue, but getting voice and video chat with yahoo and google chat were a big pain. I couldn’t even get a satisfactory result for voice chat alone after trying hodge-podge of plugins and programs, and it felt like a big mess. Even when I got the voice working, it would often stop working for unknown reasons, and troubleshooting was a huge pain. I do development work with vim on GNU screen (whether linux or cygwin), but when it comes to setting up multimedia, it was a very unpleasant experience.

On a different note, I and my friend recently tried to use Ubuntu as the OS for the HTPC setup with XBMC. Again, it was very frustrating when the fullscreen mode didn’t work, and there were misc. issues with audio.

In summary, while I fully enjoy the command-line, I hate the the GUI interface and the multimedia experience. The later may be because most of the hardware companies don’t support linux and provide drivers. May be chrome OS has the potential to change the scene for Linux in general.

Just convince Activision, EA Games, etc to port their games to linux and you will see the market share increase, meanwhile we are forced to use Windows or at least dual boot.

Maybe sites like facebook or another big site.. google.com should have a survey where the users can vote for their favorite OS (with comments optional to know why) instead to try to use automatic tools or js scripts.

For ther hand who cares, with the current dev tools all the programs and games should be cross platform.

I’ve no problem with the conclusion that Linux on some markets have less than 1% market share, meaning it has less than 1% hits on selected sites that supposedly represents, if true, all demographics in one geographic region. The interpretation is up to everyone looking at it. I’ve absolutely no problem when folks say “Linux sucks”, for whatever objective or subjectiv reason.

What I can’t understand though is how technology sites continues to show ignorance of history, like they never read computer history properly. “Linux has been around for almost two decades now” as a kernel, and when it did hit 2.0 stable you could say it had proved itself to have a future as a kernel. Where not counting a kernel, that means a layer between hardware and software, as a full fledged desktop system, are we? Linux as an alternative desktop system hasn’t been around for the general public that long, and wasn’t easy to install pre 21th century, hence limited to pretty initiated individuals. 1998 or 1999, I don’t remember, IBM chose to support the kernel development, something that gave Linux a boost, but not on the desktop. At the same time we see the first attempts to present a consumer product for the desktop. This is the time when Windows at the beginning of 2000 or 2001 reaches its all time peak of desktop OS market share at about 98%, a time when Microsoft controlled most consumer options, could engage in battles we now view as legally questionable to say the least. Windows was about 15 years ahead in its development cycle, and because of the infancy of the market it had been able to cash in huge sums of money and control the PC infrastructure. The statement “I’m doing a (free) operating system (just a hobby, won’t be big and professional like gnu)…” doesn’t equal a competitive desktop system with enough power to challenge the big players.

What we’ve seen is an enormous impact of Linux and the open-source model as a totality. 1% of a claimed market is far from the most interesting story here, and the greatest battles won haven’t taken place on the desktop, but they all have and will have a great impact on how we compute. This success has gained Windows users as well.

What’s under the heading of “What can be done to grow Linux on the desktop?” shows a lack of understanding about the fundamental mechanism of how open-source evolves. It’s dead the day you decide to control it. Market share doesn’t even matter the same way as it does for proprietary software. Among the ones I know use Linux every 10th is a techie, and every second is below an average user, who probably never will understand computing beyond habitual mouse clicks. Damn right, Linux has to improve, but killing its life stream won’t accomplish that. Why is it wrong to have differentiating desktop systems? We already see the light of Apple style closed Linux systems, Android, which probably will give you just that: one size fits all.

1% is represents about 2 150 000 sold PCs from the release of Windows 7, and that’s less than a year. We’ve only passed the one billion mark of personal computers worldwide, that means 10 million Linux PCs. In all other markets of devices of this magnitude that’s an enormous number, but for PCs it seems to counted as nothing. If most users, as claimed by this article, are techies it must be the desktop with most brain power under the hood. If so I have to agree with your statement that “Linux is not in any danger of going away”, but it might not become exactly what you expect it to become.

When you write about “getting manufacturers to include Linux as a viable option on their computers”, it’s like you’re living on another planet, a nice one where every company respects the other and willingly share room. Does the computer business work that way? Even on the hardware level it has been an ongoing dirty battles. Proprietary software companies are nice though and doesn’t interfere when the little guys try to get room, do they? I’m sorry, but I can’t resist being ironic when folks writes pretty fairytales undoing our history.

To conclude: yes “perhaps Linux will just continue to be a small niche desktop OS”, meaning that 10 million use it today and another 10 million in less than five years (that’s the prediction for PCs). As long as it doesn’t become yet another MacOS or Windows that’s just fine, and gives it a solid market injecting energy into its core of developers.

The only thing Linux really needs is a marketing campaign.
This will likely never happen, as Linux is free. To adopt a website (Wikipedia) or a browser (Firefox) is one thing, but to:

1. Download an ISO file in the hundreds of MBs.
2. Use a blank CD, and write the image to it.
3. Figure out how to boot from the CD.

For a novice PC user, these steps present a fairly large amount of time and effort.
Wubi and Mint4Win go a long way towards eliminating this barrier, but they’re far from perfect. Sadly, Wubi doesn’t even work with Windows 7.

I think a further issue hindering widespread Linux adoption is that it doesn’t solve a problem. Usability, focused GUI and all that is nice – but you still need a reason for existence. Outside of a small cult following there is no reason to use Linux.

I would like to know what “Other” is because it’s kicking Linux a$$ets.

Don’t care if Linux is on only 1% (or 10% or .1%) of all desktops…it is on 100% of mine!

In my humble opinion, a big problem is support, for those who are not “geeks” who thrive on technical challenges.

In my experience, the forums are quite helpful with newbies. But for those of who have gotten their feet wet but are by no means “experts,” support can be difficult. If one asks a question the linux gurus consider to be googlable, they’ll flat out ignore you, if not insult you. If you post a question in the wrong forum, you may be chastised.

If the gurus who habituate the forums determine that, judging by the technical content of your request for help, that you would, in the long run, be better off by using your own skills to work through the problem instead of having the benefit of an actual helping hand, they will let you dangle until, after four days of desktop R+D (never mind your actual objective), you wrangle a solution by brute force, if necessary. And you learn.

But not everybody has time for that.

The plain simple fact is the average user is a simple as a sack of nails and does not want a product that makes them think. Thats exactly what Linux is, an OS that requires some thought to use. This is why Microsoft’s platforms (XP / 7) have so much acclaim. Also Linux still is not compatible with alot of mainstream software available today which will always hinder instead of help.

We own a firm in macedonia the first and only one specialized for linux distros.ask yourself why is linux best os for servers,it has a great stability,low cost and is extreamly safe so its used by the largest and greatest states,goverments,corporation etc…
wait until the one and only google makes this new os based on linux ,os that will have best linux code.it will be an extreame shock for all os that is leading the market now soo we say that linux will be in every table top os from its release by GOOGLE.

No out of box voice chat support from msn, yahoo and google was also the decision making point for me.

At least for the web enthousiast community, Linux is growing as a desktop OS. w3cschools for example shows a slowely but steadily increasing adoption. It’s now at somewhere around 4.5% or so. Now this is indeed a technical growd, but definitely not hardcore technicians.

That stat is almost certainly under-reporting. Take for example the Brazilian Linux desktop they rolled out to most of their schools, that’s 10′s of millions of desktops none of which I bet gets logged by the websites using StatsCounter.

In my opinion what the article does is that it mixes “usage” and “market share”. It may be true that market share is low for linux (I am not qualified to say if the statistics are under estimated). And it might be also true that market share is most interesting to companies willing to sell Linux. However, focusing only on market share is not very interesting when talking about free software in my opinion. And companies producing software for the different platforms should rather be aware of their user base than the amount of people that actually purchased services from Red Hat or Canonical.

What a load this all is! No one will ever be able to count how many PC/laptops run linux because most people just install them. I have 30 families in my community that have a total of 50 computers running ubuntu. All came with MS crapware. Where are they on those charts? This isw all garbage from this blogger. Anyone who really knows linux knows that facts. LINUX people dont care about your dumb desktop share % go blow hard for microsoft and be spoon feed the rest of your life.

Its quite a safe bet that Linux on the desktop is going to start growing more rapidly from now on. Main reason for this is that more and more applications are moving to live on servers on the internet. The only thing needed on the desktop to run them is a modern browser. Both Mozilla Firefox and Google Chrome are available for Linux.

Then there is Google Chrome OS that will be built using Linux. OK, not for desktops to start with, but it will pave the way for creating attention for Linux on personal computers.

I also believe that both companies and the public sector would welcome the idea of using a better controlled environment on what their staff can do on computers. Having all the applications they need for running their business/organisation server based and used through a browser will without a doubt equal to less costs; lower hardware costs since less powerful computers are needed, less maintenance and administration since the OS can be slimmed to the minimum needed, more secure compared to Windows, etc..

Basically, in a couple of years the OS on the client computer will be less important as long as the internet clients, browser, VOIP, etc, exists for it.

How big percentage of Android owners even know its build on Linux for example? Maybe some have heard about it, but do they really know what it is?

/thomas

Who cares about desktop market share? The desktop wars were over before Linux was born. Linux dominates the platforms of the future. In the mobile space Android has already passed the iPhone, in five years Android/Chrome will be as dominant in the mobile space as Windows is in the desktop space. In the cloud space Linux is already completely dominant, all of the players with the single exception of Microsoft use Linux as their platform.

The term Desktop is highly misleading when it comes to Linux distros. Linux distros give the individual user the combined capabilities of enterprise and HPC servers as well as the functions traditionally associated with desktops. The value of using Linux is that it’s so flexible and extensible. If you want virtual machines, it’s there, clustering, it’s there, the ability to run on multiple machines simultaneously, it’s there. I generally use a half a dozen systems simultaneously, I could never do that with Windows.

The reason that Linux has a 1% share of the so called Desktop market is because that’s the percentage of the population that has the skills to be able to take advantage of added capabilities of Linux. If you don’t have those skills then there is no point in using Linux as a replacement for Windows or Mac.

As if it matters really, Linux touches the lives of nearly every single person every day. It may not be in their desktop but it is in their pocket, car, tv favorite web sites etc. I see the smart phone as the ultimate battle ground right now and Linux is starting to win that as well.

The Linux desktop share is without a doubt small, but anyone with half a percentage point of wit knows that there is no reliable way to measure the distribution of various OS’s, and in particular if depending only on data on computers that go online, and only from a selection of websites (3 million tho they may be).

These numbers are inaccurate and unreliable, which means they could be lower or higher. Now let’s all talk about how to promote Linux on the desktop, but let’s not base our effort on something like Statcounter’s data. Please.

I think most Linux users just don’t care about what other people use on their desktop, me included.

I used Linux as my main desktop for 10+ years. Today I use MacOS. Why? Because the Linux desktop was actually getting less usable. Developers concentrate on reinventing the same basic features over and over and over without making any real progress on stability or usability. Basic features like audio are still troublesome. The interface is still just a bad mix of Mac OS, Windows, and X.

The most interesting graphical Linux is Android and unfortunately it suffers many of the same problems. It isn’t consistent and it’s usability sucks. It looks like it was designed by a bunch of geeks with no idea how to design an interface. Same old story.

iOS is the best and most innovative graphical Unix environment. It had the balls to reinvent how a graphical environment should look and it is attractive, stable, and easy to use. If the Linux desktop, or Android, wants to succeed it should learn from iOS and copy these points. Attractive, stable, easy to use. Rinse and repeat and don’t be afraid to shake the boat.

The good news is that things are changing. It’s not to late for the Linux Desktop. We just need a leader to envision the future and guide other developers on the way.

imho, the main problem is that multimedia does not work outta the box.
Newbies are forced to choose between ‘ugly’ sounding illegal pkg’s or fluendo. People are given explanations about policy of everything being open and free.
Hey man! All I want is this mp3 or wmv, or whatever video, dvd, etc.. to run.
And god forbid if I try a few different multimedia apps, and/or engines.
Do you realize how many options these apps have in order to get somthing to run? Yes, I know, some are lucky and one app will serve their needs. But many are not. I have happily used Linux since Fedora 2, many distros, and have quite a bit of experience with this subject (and many forum threads with big wigs to prove it).
I would say to rh/suse/debian, it is worth it for you to do anything to ensure that at least 90% of all codecs, including browser plugins, just works upon FIRST encounter. Even if it means that YOU pay some royalties behind the scenes.
I see that the MINT distro is going in that direction but I have not tried it yet and don’t know if its as simple as multimedia on Windows .

Quoted from Rainer Weikusat in the comments on Linux Today from this article:

“According to Netcraft, there are roughly 96,000,000
active sites on the web (rounded up). StatCounter claims
that its tracking software is installed on ‘more than 3
million sites’ which means ‘less than 4 million’ and
probably ‘a lot less than 4 million’. Consequently
StatCounter gets information from less than 4.1% of
these active sites and from more than 3.1%, IOW, traffic
going to at least 95.9% of the existing active web sites
has been completely ignored for the purpose of this
‘statistic’. ”

Having been in the “stats” game, I know that statistics can be bent and molded to serve any purpose by anyone. Weikusat illustrates a glaring hole in this presentation and fairly well provides enough empirical evidence to discount the numbers.

As a small non profit that has installed Linux on over 1100 computers in five years, I am left to wonder who has counted these numbers. Multiply that by millions who have installed Linux on computers for their parents, children or friends and you are left to again wonder at the accuracy of the article.

I am immediately suspicious of any article or publication that relies on specific statistics to prove their point. And while not particularly aligning myself with any particular distro, I too have dealt with problematic installs of skype and MSN/Yahoo chat. In Ubuntu, I faced these issues often. When doing the same installs in Linux Mint, The setup was straight forward and I had the software working in minutes.

While I don’t have any “statistics” to back up my theory, I am guessing that blind fanboi-ism has done more to hurt our efforts than most anything…

I’m just sayin’.

Rarely does one read an article so far off the mark. Letting alone for the moment the questionable methodolgy of the figures….one should ask about the unbelieveable number of businesses running IE6 or Win XP SP2. Are they doing that because Linux is hard to learn or there are several versions available? Or is it because IT managers are incompetent,clueless or at best, powerless?
For the average home user, yes they would rather kick an extra hundred dollars or more to MSFT rather than being forced to think – even a little bit.

As long as Linux is on one desktop – i.e. mine, I’m quite happy.

I’m no techie but I disagree totally with the first two points. Choice of distribution and GUI are a big part of the attraction of Linux for me and I wouldn’t care to lose them just to attract a bunch of (probably unappreciative) Windows users.

Didn’t they try something like this with Linspire. I seem to remember that neither Windows or Linux users really took to it.

This article is pretty pointless. You know what would make ice cream better? Ice cream would be better if it were all one flavor. Mint Chocolate Chip is great so if all ice cream were Mint Chocolate Chip, then ice cream would sell better. That’s basically the argument made here. The great thing about Linux which we should be championing is choice and freedom.

The needs of a business user are not the same as the needs of a developer or a home user or a gamer or a graphic artist. Linux meets all those needs by not being the same out-of-the-box cookie cutter OS that gives you no choice or freedom.

Embrace the KDE versus GNOME battle as a good thing because the competition will make each project better. If you want something that looks the same and works the same and doesn’t have any desire to innovate, stay on Windows.

I understand Linux needs to be easier to use however Windows seems to have evolved into a mode of thinking that the user is a complete idiot. The flexibility of a Linux system to do pretty much whatever the user wants is a strong reason to chose Linux in the first place. I’d hate to think that attribute would be sacrificed for the sake of ease.

WHAT YOU WANT IS CALLED MINT.

Amazing. Netapplications, a web business which markets rebranded Windows applications tried this stunt a couple years ago. StatCounter “appears” to be neutral in publishing statistics nearly identical to Netapplications’. When one uses Netcraft’s site report on the domain name they are presented with Panther Express Corp (93.188.128.22), which is NOT the IP address of http://www.statcounter.com. Using other tools on 66.114.48.10 we find

NetRange: 66.114.48.0 – 66.114.63.255
CIDR: 66.114.48.0/20
OriginAS: AS40366, AS36408
NetName: CDNET-USA-2
NetHandle: NET-66-114-48-0-1
Parent: NET-66-0-0-0-0
NetType: Direct Assignment
NameServer: NS1.PANTHERCDN.COM
NameServer: NS2.PANTHERCDN.COM
RegDate: 2007-07-05
Updated: 2009-11-10
Ref: http://whois.arin.net/rest/net/NET-66-114-48-0-1

OrgName: CDNetworks Inc.
OrgId: CDNET
Address: 130 Rio Robles
City: San Jose
StateProv: CA
PostalCode: 95134
Country: US
RegDate: 2007-01-05
Updated: 2010-05-10
Ref: http://whois.arin.net/rest/org/CDNET

StatCounter is CDNet’s project and their bread is buttered by Microsoft. They stopped offering Linux application downloads through their site around 2007. Doesn’t look unbiased to me.

Steve Ballmer gave a speach in Feb or 2009 during which he presented a graphic which showed the then current desktop market share. At the time, Apple’s market share was around 10% (retail channel counts trump web site pixel bots) and the graphic showed the Linux share as being slightly larger than Apple’s slice. So, who is lying, StatCounter, Ballmer or both?

What surprises me is how a Penquin could be so easily duped by YALS (yet another lying statistic).

Stock markets around the world are moving to Linux because of its speed, stability and security. The latest is the London Stock Exchange, which is moving to a Linux trading application (they even bought the company that wrote it) after suffering a $1 Billion dollar loss because of the failure of a .NET application written by Microsoft and one of its partners in England. If the folks who own the API and know of its undocumented features can’t write software that won’t fall over under pressure, who can? Militaries and their governments are moving to Linux.

There are more people running Linux in China than there are people using computers in the USA, and quite a few in India. Even Hollywood and Bollywood use Linux servers to create their computer generated movies. Google, Amazon and many, many other major retailers have grown tired of the financial costs of trying to keep Windows secure and now serve the products and are switching to Windows on their clerical desktops.

Microsoft is setting on a huge pile of cash, most of it the fruits of an illegal monopoly which gouged consumers while paying PC OEMs to keep competitors off their desktops, just like Intel paid DELL not to use AMD’s graphic chips (Intel’s bribes amounted to 75% of Dell’s income in 2007). Even a casual search off the Internet will find stories of reporters who learned, after the fact, that they were groomed to write stories which amounted to MS PR memos. Bloggers were given gifts of expensive laptops to write favorable stories about VISTA. To anyone who was read the Comes vs Microsoft 3096 PDF this kind of “slog”, “stacked panel”, compromised analyst or consultant, etc., is nothing new. Now, with two examples as evidence, we can add the “stacked statistic” to the list of techniques MS and its “Technical Evangelists” use and for which James Palmondon has performed a mea culpa and appologized.

This CDNet supported website is just another tawdry example.

Yes, desktop Linux may not be breaking 1% on your choice of source for stats.

“Sadly”, desktop Linux has “only” managed to claim 100% of Brazillian public schools (52 million + students), penetration into some 10% of government and business offices in Germany, significant deployments into several regions of spain, multi-million retail sales per year (think: local “big box” style stores) in both South America and Asia … and on and on.

.. those probably aren’t going to show up on your stats, but the world is bigger and rather more interesting than StatCounter.

Desktop Linux is doing a lot better than the chicken little stats people love to trot out on blogs like this.

That said, I do agree with points 3 and 5 in the authors “list of things to do”, particularly point 5 about target audience.

The rest of the list of suggestions hardly add value in terms of bridging real world barriers to entry. Point 4 is useful and should be striven for, but it’s not a major key to success; it is the old “better technology wins!” falsehood masquerading as a wise point in a list. The other two points about standardizing into a monoculture are unrealistic and unnecessary; they also represent a gross over simplification of the actual landscape as it isn’t a set of diametrically opposed, cleanly separate solutions but a set of interdependent and symbiotically developed offerings.

The biggest thing the author missed, though, is the most potent one: get Linux pre-installed and selling on laptops, netbook and desktops through broadly and easily accessible distribution channels. That is the #1 head wind we face. In areas where desktop Linux is popular, such as Brazil, this has already been accomplished.

Linux is for those who don’t mind occasional mental effort. Most people (in the US, anyway) want an OS that “just works”. It’s a problem with US culture, which has become profoundly anti-intellectual in the last 20 years. It won’t change until US society gets over its current lunacy.

I’ve tried to make the switch more than once to a Linux desktop. I’ve used several distros to try and accomplish this. The biggest issue I have, however, is the lack of consistent VIDEO DRIVERS!!!

I have a middle of the road, not so old card… and I either have to jump through hoops to get it working or when it does work out of the box, I get very poor performance. Now, I don’t run games or anything major. I usually just want to watch hulu or a video here and there but with Linux doing any of that while having some desktop effect, just won’t happen. I’ve changed cards to a new one, thinking that it would work better. It doesn’t. Since it’s an NVIDIA as opposed to an ATI, now it becomes very difficult to get dual monitors.

With Windows, it works. It does not matter if it’s WinXP, Vista or 7… Video drivers just work.

This article misses the point entirely.

1. Linux isn’t trying to become successful on the desktop and
2. Without choice, Linux doesn’t exist.

All you people want to do is turn Linux into Windows. There can never be one unified anything with Linux, that’s kinda the point.

This is just another article by someone who totally and utterly fails to understand what Linux and FOSS are.

These statistics are representative for the U.S. It should be noted that Linux isn’t as marginal in Europe. Especially Germany is rather technic-literate with a market saturation guesstimated between 2 and 3%.

Funny thing is. Everytime a pro MS partner comes along and posts on of these articles. Providing an illegitimate place for MS shills to go on and on and on about usability and install base and this and that. The one thing they are actually doing, is pushing more and more people toward Linux. Linux gets better daily, it wipes the floor with Microsoft, and they know it.

Microsoft will remain the dominant OS right up until the point it ceases to exists in the minds of people that write these articles.

Right now MS are flailing about, with less and less of that corporate dollar going into their pockers. OEM’s care less and less for them, now offering choices on purchased machines. They know the gravy train has run dry.

With Steve Ballmer at the helm the company has gone down hill rapidly, showing it’s true nature in public often.

I don’t know why they just can’t face facts. It’s over for them, they had their run, now the world wants freedom. Whether they like it or not, people are taking that freedom, from colleagues, friends, acquaintances, or off their own bat.

The installed base of Linux is massive and it got that way without a shred of advertising, no illegal methods, and a monolith of finance fighting it at every turn.

Windows could never have one against a global awakening of the end user. So just give it up with these repetitive, tiresome and totally non-factual rants. It’s pointless.

There’s a not particular recent picture around showing Steve Ballmer at a presentation with a market share graphic. The graphic shows Linux at about 5%.

MS counts every Windows unit sold (by unique serial number at least).
MS records every Windows installation authenticated (or activated)
MS records (in later versions at least) will show how many of those active copies of Windows cease to be active.

The greater majority of those inactive copies of Windows will be Linux, and MS can make an intelligent estimate of hardware die-off. So if Steve says about 5% that is probably a better estimate than any one else can make.

Until computers leave the factory with Linux, Linux adoption will be less that 5%

I don’t know about you, but the main issue with Linux is not Linux itself, but the rest of the computer-oriented ecosystem. Read: continous hardware drivers available.

In 2010, you STILL have the risk of purchasing a printer/scanner and not having it to work 100% like in a Windows or MacOS X environment. I believe that is even worse than the lack of gaming or killer apps.

If I could be sure about getting ANY piece of hardware, no matter how strange and misterious it could be, with a native Linux driver and that will behave EXACTLY the same as in “more mainstream” OS(Win/Mac) then the Linux desktop would rise a bit.

Using Linux it’s still a bit of a risky business I guess.

I note a remarkable similarity to my post:

http://www.happyassassin.net/2010/07/29/the-success-of-ubuntu/

hmm. =)

For those banging on about multimedia, e.g.:

“I would say to rh/suse/debian, it is worth it for you to do anything to ensure that at least 90% of all codecs, including browser plugins, just works upon FIRST encounter. Even if it means that YOU pay some royalties behind the scenes.
I see that the MINT distro is going in that direction but I have not tried it yet and don’t know if its as simple as multimedia on Windows .”

It’s just not that simple. It’s not legally possible to pay a license fee to cover the distribution of a F/OSS decoder or encoder for a patent-protected media codec, because the F/OSS license requires that the code be redistributable to anyone in the world, so the patent license would have to cover…everyone in the world. And most patent holders aren’t willing to grant such licenses. As long as we want our distros to be redistributable by people who download them, and we do, patent licensing isn’t an option. (Note that Canonical *has* dipped a toe in the waters of patent licensing, using closed source codecs, for OEM installs…but that’s the only way you can do it, and it means that if you buy such an Ubuntu preload, you can’t technically redistribute that copy of Ubuntu to anyone else, at least without stripping out the licensed codecs).

We’d love to include patented codecs in our distros, we really would. But it just isn’t practical. If you’re wondering about Mint and similar distros – simply put, what they do is not actually legal, if they provide their distros to the U.S. But there’d be no point suing the developers of Mint, because it’s a community project and has no money; you’d just lose a bunch of legal fees. So, practically speaking, they can do it and not worry. Distros backed by companies with money cannot, though.

aseigo: “.. those probably aren’t going to show up on your stats, but the world is bigger and rather more interesting than StatCounter.”

Why? Do people in schools and government offices not browse the internet? That would contradict all the real-world experience I’ve ever had.

Actually I think those numbers are perfectly consistent with reality. A lot of Linux proponents really aren’t used to thinking *big* enough. All Brazilian schools? That’s great. How many countries are there in the world? A couple of hundred. How many of them have decided to use Linux for all their schools? Well, one. There’s 0.5% right there. I rather think that all the deployment numbers you cite are probably right around 0-1% of the total available user base in those sectors.

“Desktop Linux is doing a lot better than the chicken little stats people love to trot out on blogs like this.”

I really haven’t seen any evidence of this. Believing otherwise requires believing that there’s a truly gigantic amount of desktop Linux users out there who somehow manage to permanently dodge all the web usage stat tracking systems in existence. Why would you believe that? Is there any evidence at all for the existence of such people? If they’re not browsing hugely popular web sites, what *are* they doing with their desktop systems?

Wow… A ton of interesting feedback. Thank you everyone for your comments, opinions and thoughts.

A couple of the commenters seem think we are “pro MS” for some reason. The truth is, though, that we tried to approach the article from a neutral perspective. (Here in our office we run a mix of Linux, Mac OS X and Windows.)

Linux is neither a product nor a company. It is a product of a complex network of interests and relationships. The most likely result of any drastic change, be it proposed to help Linux to gain more market share or not, is the total elimination of Linux through destruction of its ecosystem.

The right way to make the Linux desktop share larger is to change people, not Linux.

I guess the recent success of tablets will increase the Linux desktop share since pure content consumers will use tablets and free the desktop of their presence.

Stats for the world show almost 5 percent Linux…
http://www.w3schools.com/browsers/browsers_os.asp

Paul Cobbaut: w3schools gives a Linux number consistently higher than all other surveys. The generally accepted explanation is that the audience of w3schools is significantly skewed in a geekwards direction.

Wikipedia keeps a survey-of-surveys:

http://en.wikipedia.org/wiki/Usage_share_of_operating_systems

you can see that w3schools is definitely at the high end. The Wikimedia numbers may be interesting to some commenters: is the hidden, huge group of Linux users not visiting Wikipedia? Do students not use Wikipedia?

Haha, these comments were not written by marketers…
That is all I need to say, and also applies to how linux is perceived, and will continue to be perceived until someone takes the reigns somewhere and gives Linux an image that spells ‘usability for all’ as opposed to ‘linux, yeah I heard of it, you run servers off it? I’ve wanted to install it but I don’t have the time, and it wont do anything for me, and I don’t know how. oh well.’ There needs to be a big shift in perception, but also an opportunity to take the leap, if possibly at no risk, and minimal effort.

Yeah but it’s the top two percent. The cream of the crop. :)

“if possibly at no risk, and minimal effort”

Oh god.. What hope does Linux have with people thinking like that. No effort. It’s an ENTIRELY DIFFERENT computing model. No effort…

Read http://linux.oneandoneis2.org/LNW.htm

and to the guy up there that said Printers aren’t as compatible with Linux as with OSX? Do you know where cups came from? (Common Unix Printing System). The printing system used in Mac OSX and Linux…

The title of this article says how biased you are: “The consistent failure of Linux”.

You could have equally have said “The consistent growth of Linux”… Even though it may not be a market share giant it’s year on year growth has been increasing rapidly.

These numbers are plain WRONG.
As showed even on a Microsoft presentation, Linux already has a bigger desktop market share than OS X.

Don’t trust any statistic you didn’t made yourself :)

The linked website is down, so I don’t know, how the statistics was done.

The reason for low usage of Linux (as I see it) is mainly the rigid environment of big companies and pre-installed computers. In my country (central Europe), I cannot buy laptop without windoze. So at the end I have 3 Linuxes, but they count as (are sold with) windoze. Such statistics are useless.

Fortunately, youngsters have today very broad experience with Linux (Mac is uncommon here), roughly tens percents prefer it.

I use PCLinuxOS. It’s quicker to install than Windows, does everything I need it to do (but then I’m not a gamer), and runs FAST on a 4-year old machine. It didn’t cost me anything. My printer works, scanner works. No problems connecting music players. No problems connecting external storage. Some problems connecting mobile phones but not insurmountable.

No driver disks required, all hardware detected. Stable. No virus worries. Internet connection good. I set up a dual boot on my Samsung NC10 netbook with Windows XP, couple of tweaks to get the brightness up/down buttons working, everything else out of the box. Shared the tweaks on the PCLinuxOS forum ‘coz that’s what us Linux guys do.

I’m always hearing of problems with Windows. People throw computers away when a re-install might remove viruses/other stuff that slows it down to a crawl.

Linux share less than 1%? I don’t care, frankly. It’s certainly greater than 1% among the people I know – because they see how good it is and let me install it for them – and never return with problems. It’s there for those that want it. As for everyone else, well that’s their problem.

Number of households with desktop PCs by 2014 estimated to be 2 billion. 1% of 2 billion = 20 million. 20 million linux machines is a niche market?

The arrogance of too many Linux users is mind boggling. I switched from Windows about 18 months ago, did a dual boot for a while, and now run Xubuntu only (and UNE on my Asus). Both are great OSes, and I would never go back (and it wasn’t easy to switch, since I’m a freelance writer and need something like Quicken and Windows Live Writer, which really don’t have Linux equivalents). If nothing else, I don’t have to reboot anymore. In fact, I’m the kind of computer user that Linux needs to get if it will ever make any inroads on the desktop.

But to hear the Linux fan boys talk, I’m a sellout. Xubuntu is bloated and UNE is too slick. I need to run Slackware from the command line, or I’m a weeny. This is ridiculous, and the Linux cause will never be advanced as long as people think that way. How shallow and selfish is that approach? It’s as bad as the Mac and Windows people, frankly. If a product is better, the last thing you should want to do is keep it a secret.

from what I know, the main problem with Linux is the lack of office tools like Mi$soft Office. I showed to alot of people the Linux interface, and it’s features, but when the question was asked:”How obut the Office equivalent?” and they saw the “sad” Open Office the answer was “I’ll stick with Windows for now.” You have to admit that a regular user uses 80% of his time the Office Pack and in the rest some workplace software.

I think this is the first place where Linux should be improved. It the Open Office will look like the Gnome interface the adoption figures will look very different.

Bottom line is that it needs to be taught in schools and colleges for people to feel ‘comfortable’ with it. Making it shiny isn’t what’s needed.

oh and a proper outlook equivalent would help

“And this may be the hardest one of all: Don’t build for Linux’s current hardcore audience of tech geeks. If the goal is to bring in new users, that approach is a big no-no since it will alienate the vast majority of the population”

1. If you want more people to use Linux, than advertise it.
2. You don’t need to dumb things down cause some PC illiterate doesn’t like it, they can use a mac.
3. Linux is a practical OS that does not need stupid shit in lieu of a great interface. If its too hard for you than learn to break through the learning curve.

@Johnsie, for the mail part you don’t see thunderbird as a proper equivalent to outlook ? Web browser and mail client were the easiest to change when i moved to linux. I agree with @Racoon, though…

I’ve been using Linux on and off since 1998 when I installed Red Hat for the first time. Ever since I’ve been amazed over how things are better than Windows and how cynic the Linux community are. The complete failure on desktop OS is a result in it’s own strenght: choices! Sometimes people and developers does not want to choose desktop, drivers etc. What they want is a system that works! Windows is not the best but it works for the task. The biggest failure is to get serious software developers interested. If Linux shall compete with Windows and OS X it’s not enough to be “equally good”, it has to be better! The best Linux software today such as Gimp, OpenOffice, NetBeans etc can be run in Windows just as good or better. The opposite is not possible so therefore the choice is easy. I have a lot of bad things to say about Windows but also good things. The best is the huge software library that supports the platform and that is Linux biggest problem.

Just a thought: more and more applications are migrating to the web. Soon, Google will release a lightweight operating system that will act as a gateway to Google’s online products. In 10 years, operating systems make become obsolete, or nearly obsolete, existing merely as thin bits of firmware that connect your device to the cloud (many people even expect games to be played through browsers in the future, which will solve the problem of “no games on Linux”). So “desktop Linux” won’t matter, and neither will Windows. Server operating systems will be everyone’s operating systems, and that’s where Linux is already thriving.

lol i gave to say this so Much those statics are wrong linux has about 2% more then mac

Linux will most likely take over the PC gaming market i think it will with SteamOS around the corner Linux will win linux is the future i said this for a long time it will happen once we get more applications for linux i try send to the developers to make linux versions of some softwares and most of them have (its almost like i work at the liux foundation getting people to make and promote linux software0
for example at the moment i am trying to get piriform to make a ccleaner version for linux and some other ones…

@Johnsie There already is it is called Thunderbird made By Mozilla a Proud Supporter of Open Source Software!!!!! :)

oh see how quickly linux is evolving more and more software all the time

@thelynx tell us why linux is what?

tell us why linux is what?