OneStat today reported that Microsoft’s Windows dominates the operating system market with a global usage share of 96.97 percent. The Mac appears to be at around 2.47% and Linux at 0.36%. Please note that OneStat bases its numbers on a large number of internet site hits and extraction from user agents.
Microsoft Windows Dominates the OS Market on the Web
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
In that case, should we take the statistics with a grain of salt?! I know I do!
These statistics are incomplete, but than again, numbers are numbers and this is how big software makers justify their dominance.
Thoughts anyone?
I would like to see Google’s statistics on the subject. Google is the only website I can think of that would get hits from people of all walks of life and wouldn’t be subject to content bias.
Remember, it’s not the default search engine on most people’s PCs. You need a certain amount of technical know how to change that.
Yes, but it is by far the most widely used search engine in the world. Yahoo! is number 2, with approximately half of Google’s marketshare.
True.
However, consider this. My mother knows something about computers, but her use is limited to office programs and very occasional web browsing. My father knows less, has trouble understanding the file system, and uses the Internet mainly for paying bills and sometimes browsing the pages of travel agencies.
Yet they both know and use Google. In fact, I don’t think either of them even knows there are other search engines. It’s only partly my influence, as I have showed them a couple of alternatives and eventually set up Google as Firefox’ homepage. Before this happened (in 2004, I think), they had been aware of Google for a year or two, as all their friends and coworkers use it!
I don’t know what the situation is in other parts of the world, but from what I’ve seen, Google is by far the most used search engine in Finland. If dogs used computers, I could say “everyone and his dog uses Google”. It’s used by computer-savvy and non-savvy people alike.
Exactly. I haven’t at any point in the last 3 or 4 years seen a single person pop open a browser to go to msn.com or yahoo.com to do a search, everyone around me uses Google. I live in Canada.
Not to mention they could track it on a per-user basis as they supposedly track their searches that way anyway!
If that’s true, then I personally know practically every Linux user on the planet. Somehow, I have my doubts…
Well if we assume that there are roughly 1 billion people using the Internet (a very rough assumption I know), you personally know around 3 million people which is bloody impressive!
says “Proud User of GNU/Linux”
this statistics can’t be true!
More Intel Mac users as PPC mac Users? 4x more?
It’s a dishonor to see that MS has that big market share! Every little child know that monopols are very dangerous! It’s like a dictatorship in the market.
More Intel Mac users as PPC mac Users? 4x more?
According to this article, it is actually 15x more, which is laughable.
I somehow doubt you know 7200 Linux users, let alone the 7 million worldwide this survey indicates they are (it’s called multiplication buddy). You must be very popular!
OneStat doesn’t say what sites they’re monitoring, only that the data comes from a pool of 2 million visitors. So they could be monitoring Windows-centric sites, or sites whose clientele is required to use Windows, or sites that host Windows-only content. These stats are bullshit — 0.36% of 2 million is 7200. Are they trying to tell us that there are only 7200 Linux users worldwide?
Edited 2006-08-14 17:12
These stats are bullshit — 0.36% of 2 million is 7200. Are they trying to tell us that there are only 7200 Linux users worldwide?
No, that would mean that among those 2 million visitors, there were 7200 Linux users. 2 million is a small snapshot of the total internetting population.
Also take note that the survey reflects people who use Linux on the desktop in leisure time to surf the internet, thus excludes those who use it on the server, graphics editing department, embedded devices or whatever. It’s indicative of Linux on the desktop, not Linux in general.
Of course I can’t judge their methods (sites used etc), the site appears to be down at the moment.
Um, no they are not claiming there are 7200 Linux users worldwide. I assume from your logic that you failed Maths.
OneStat claim that 7200 of the 2 million visitors they logged use Linux. They are most definitely not suggesting that there are only 2 million people using the Internet in the world.
Find 0.36% of how many people use the Internet (I would hazard a guess at around 1 or 2 billion) and then you’ll have how many Linux users there are worldwide according to this survey – roughly 3 to 7 million.
Of course personally, I think the numbers probably aren’t true. I would say that Linux has higher share, and Macs ever so slightly more, but hey it’s their survey.
Well, as long as the linux share is below 50%, the number will not be enough
The number will not be enough for …. what exactly?? I don’t think there is a general “Let us try and turn Linux into a monopoly” agenda hidden anywhere.
2 million is not the number of internet users worldwide, obviously. You should multiply your numbers by 1000 more or less to get the right figures. And that is about 7 million Linux users, whish is not bad.
I wish Linux would have a much bigger market share, but those numbers are consistent with my own expereince from having access to many different sites (from clients). None of them are Linux or Windows related. They’re mostly stores and some about history articles (both in English and Spanish), and Windows users are about 97%, while Linux hardly gets to 0.5% in them.
Unfortunately, even in Distrowatch the Windows users double the Linux ones:
http://distrowatch.com/awstats/#os
That’s just because they are there to trade in their old, worn out project OS for a shiney new one.
No, they mean 7200*3000, or: 21,600,000
3000 comes from 6 billion divided by 2 million. Although I realize it’s unlikely there are 6 billion people on the internet, but maybe there are almost that many desktop computers now…
Please re-read your comment 12 dozen times so you can see the error of your logic.
7200 out of 2 million doesn’t mean there are only 7200 linux users worldwide, it means it that pool of 2 million surfers, there are 7200 Linux users.
You would have to multiply from there.
Maybe the rest (0.33%) of the users are beos users =P
What about MS-DOS, FreeDOS, DR-DOS, and PTS-DOS?
Microsoft does not dominate the operating system market but only the market for _desktop_ operating systems that is just a small share of the whole operating system market. There are much more installations of iTRON than computers that are capable of booting Windows in the world.
There is some truth to what you say. Yet, for the web, it might well be true that Windows is the dominant operating system. Where your comment makes sense is in the fact that the os in most widespread use is the one that runs consumer electronics devices like DvD players, VCRs, music systems and the like. AFAIR, its author is Japanese. Perhaps others on OSnews will comment and clarify this.
Yes. I’m just tired of people taking about “the operating system” market and meaning “the desktop operating system market”. It’s like saying Boeing dominates the transportation market because they dominate the market for air transportation (don’t know if they really do).
A little bit off-topic but Airbus is very close to Boing (don’t have real numbers) and there are other “small” competitors (Embraer from Brazil, Bombardier from Canada, etc.) that have a small market as well, although “big jets” are mostly Boing, McDouglas and Airbus.
FYI Bombardier isnt canadian, in fact it started in finland as an engine company. Also in sales numbers Airbus by far outsells boeing in both numbers of planes and in total revenu as well as in profits. (Not that it matters but i thought that someone would care =P )
FYI Bombardier isnt canadian, in fact it started in finland as an engine company. Also in sales numbers Airbus by far outsells boeing in both numbers of planes and in total revenu as well as in profits. (Not that it matters but i thought that someone would care =P )
Dude, bite your tongue. Joseph-Armand Bombardier was Canadian, he invented the “modern” snowmobile which was the birth of the company (though they sold off that division a while ago).
Now they build jets and trains, but can’t seem to do so profitably since they’ve become addicted to corporate welfare, receiving billions of our overabused tax dollars in gov’t subsidies over the years.
Snowmobiles and subsidies. Short of throwing a case of Molson into the mix, you can’t get more Canadian than that.
Well, we are drifting off-topic, but Bombardier is definitely a canadian company… It was founded not very far from where I live. Furthermore, “Joseph-Armand Bombardier” might not sound english, but it doesn’t sound finnish either.
And to keep the off-topic more offtopic, why not add that Bombardier (I am still convinced that it is Canadian) brought/Merge with the NorthAmerican company “LearJet” from Jim Lear…
😉
And Windows still has a monopoly. But I still have to wonder how much of this is due to user agent spoofing. I know I’m guilty of it.
Only a few geeks spoof their user agent. The vast majority of user agents are correct.
The main reason to spoof your user agent is if you must use a web site that rejects browsers it doesn’t know about, yet works perfectly well for Firefox/Linux if it doesn’t know it’s talking to Firefox/Linux. So sometimes people change the agent to say they are on Windows instead. But I don’t think all that many people are in that situation, so it probably only has a small effect on the stats.
I for one have to spoof Konqueror to make it work in GMail.
I had to spoof as Netscape 6.2.1 on Windows 2000 in order to be able to use a potential employer’s web-based job acquisition form. (I’m using Linux and Firefox.) Took me a while to locate a good user agent string. The User Agent Switcher plugin for Firefox is quite handy. I also tried as Internet Explorer, but then the site began to throw me some ActiveX garbage and refused to work…
No, it wasn’t a job in IT or computer science. Luckily.
Makes one wonder why Mac users don’t fake browser User Agent string.
Becuase then the browser-caring sites treat your browser like one that requires different handling that it does and you get broken behavior anyway.
Part of the reason why spoofing is so rare in general.
The 10 most popular operating systems in the world on the web are:
1. Windows XP 86.80%
2. Windows 2000 6.09%
3. Windows 98 2.68%
4. Macintosh 2.32%
5. Windows ME 1.09%
6. Linux 0.36%
7. Windows NT 0.24%
8. Macintosh Power PC 0.15%
lol, even Micorsoft’s terrible OS windows me beats linux.
Yeah, but the joke’s on Windows ME users, and they don’t get it.
How bad was/is Windows ME? I have never used it and from what I have read and heard if you bought a computer with ME pre-installed it ran OK, but if you tried to upgrade your computer to ME then there was Hell to pay. Wasn’t ME a “true” 32-bit OS and that was the root of the problem, at least where the drivers were concerned?
How bad was/is Windows ME?
this might help.
http://en.wikipedia.org/wiki/Windows_Me#Criticism
A lot of the problems occured because people used their driver discs for their hardware, which installed Windows 98 and 95 drivers in ME. This is why, if you had it pre-installed (or took the time to find proper ME drivers for your hardware) it really didn’t run too bad. I know a guy who’s computer came with ME….it runs for days and days without any problems.
I have to say that switching to W2K was heaven, even if I had to reinstall all my apps since it wasn’t an “upgrade”. ME just sucked badly and Microsoft definitely doesn’t deny that now.
This is a recurring cycle in a lot of technology companies:
1998: Win98 is the pinnacle of what desktop computing can be! Win95 sucks!
2000: ME rocks, 98 was crap!
2000b: Um, sorry, ME’s crap, switch back.
2001: XP is the best, it’s based on the technology we should have based ME on!
Or Apple:
PPC is the best, Intel just ships cheap junk.
Intel is the best, PPC has been behind for years.
It’s a good sign that they’re willing to not mention all their regrets when trying to sell a product, and they keep the engineers from talking about what management wouldn’t let them do
. Some of this is a reality of life (you can’t perfect things and often need to ship things you’re not proud of to pay the bills), and some of it is the ugly side of marketing (ie, it’s blatant dishonesty).
It blue screened alot I used it.That was almost every day.Was a memory hog too.
I had a HP laptop that came pre-installed with ME. To be honest, I found it to run really good and stable. I watched alot of DVD’s, burned alot of CD’s, and played several games on it (mostly Starcraft type games), and never had problems with it crashing or getting the BSOD. However, I only ran ME for about 5-6 months before I put Win 2000 on it. I’ve heard all of the horror stories that people had with ME. I was just one of the lucky ones.
I too used it, briefly. Most if not all the stability problems I observed and that I could track down were related to drivers. If you had drivers that worked for Win98, you could install them on WinME, but their behaviour was unpredictable and many device manufacturers just never provided proper support for WinME. There may well be more to the stability issues. Only MSFT knows for sure. The irony is that WinME was supposed to cure many of the instabilities of Win98. For some users it may have worked fine, but just the opposite happened for many users.
Even though I started out my use of personal computing using a Macintosh, I somehow, can’t remember the “Macintosh Power PC” OS.
If the statistics doesn’t even get the names of the OSes right, how can we trust the figures.
Most other studies I have seen so far gives MacOS and Linux about equal figures, and they usually end up somewhere between 1 and 4%.
Even though the sample could have been large enough to give reliable information if it had been collected in a random way. According to the article this doesn’t seam to be the case.
Their sample is drawn from their customers. That alone would largely exclude home users that are not likely to subscribe to their services.
The very fact that this survey says that there are 15 times more Intel Mac users than PPC mac users browing the internet makes me doubt the accuracy of this survey. I very highly doubt Apple has moved that many of their users over to Intel Macs already.
According to this survey, all Macs combined equal 2.47% of the market…which sounds plausible. So let’s assume there are 1 billion internet users world-wide. That means that there are 24,700,000 Macs worldwide.
According to this survey, however, 23,200,000 of those Macs are Intel. When you consider that Apple sold 1.33 million Macs this last quarter, only 75% of those were Intel computers, and it was the most Macs Apple has ever sold in a quarter in the history of the company (according to Steve Jobs), there should be at best 2 million or so Intel Macs in the world, and the PPC Macs should outnumber the Intel Macs by far. Unless of course Apple is commiting fraud by not fully reporting their sales (far from plausible considering this would hurt their share price) something is wrong.
This is clearly flawed, at least when it comes to Macs. I’m sure people could demonstrate other areas where this survey is flawed, but that one I noticed right away.
“The very fact that this survey says that there are 15 times more Intel Mac users than PPC mac users browing the internet makes me doubt the accuracy of this survey.”
Those of us who are literate understand it doesn’t say any such thing.
From the useragent string, Onestat “knows” .15% are PPC. Since it doesn’t know which platform the others are, it doesn’t say. Onestat just says Macintosh. Some are PPC. Some are Intel.
And you are desperate to debunk something you can’t … so you make stuff up.
Tell us which line says Intel Macintosh?
No such line.
Well I presumed that the rest were Intel, sorry if I was mistaken. And I’m not desperate to debunk anything, I was just expressing my thoughts. No need to be a jerk. You’re not very good with people are you? A simple “no, this is what it really means” would have been sufficient. Christ, I have a neurological disorder that I was born with that affects my social skills, and I’m not even as socially retarded as you are.
Mmmm it a shame the statistics couldn’t tell us how many of the 86.8% using XP how many were using a pirate copy. I wonder 20% ?
I’m also very happy to be one of the 7.2 million Linux users rather than one of the 21.8 million ME users.
I’m also sure as Ubuntu, Suse etc get better and better the Linux market will grow.
The Windows/Other ratio is probably a _little_ bit overstated, but, +/-1%, it’s definitely in the ballpark.
However, I have a hard time believing the numbers amongst the Windows versions. This suggests that 90% of Windows users are running XP! Can that be possible? There are so many people running Win2k, and many people still have the balls (or, more likely, the ignorance) to surf the web with 98/ME. I’d have guessed that the XP penetration among Windows users would be closer to 75-80%.
There’s got to be some significant sampling bias going on here… Your thoughts?
I think it’s about right, XP has been out for so many years now (5?). Aren’t we complaining Vista is taking for ever 🙂
This suggests that 90% of Windows users are running XP!
I don’t have a hard time believing that. Virtually every home computer purchased in the last five years has shipped with Windows XP pre-installed.
There’s got to be some significant sampling bias going on here… Your thoughts?
Yeah, there’s something up with that. The stats last summer still had 2K ahead of XP, it was something like 42% vs. 37%. There’s no way it’s accelerated that much.
I think there’s two problems:
a) As mentioned, cultural- or market-bias in the sample pool of websites, without knowing the basis of the sample it’s meaningless
b) OneStat and other tracking sites are generally blocked by most third-party utilities, and with browsers like ff/opera/konq (and probably others) supporting simplified blocking or privacy controls, i’d speculate (and it is simply speculation) that a significant number of linux machines are simply not tripping the counter. Anybody blocking third party cookies, let alone using public filter lists, will not be counted.
There’s also the whole issue of spoofed useragents as well, though I’m not sure of the impact there, I’m not clear on whether spoofed clients on Mac or linux are also misrepresenting the OS.
Basically tracking stats are too inexact and flawed to be used as any sort of a reliable metric in this manner.
I don’t believe this simply for all the other stats I’ve seen in the past few years. The ones from the IDC or some other place that have consistently put usage at 2-4%. The same ones that say that GNU/Linux use is roughly the same as Mac use. In the years I’ve been using it (only 3), this is the first stat I’ve seen that puts usage below 1%.
Here’s the first non-Linux magazine link I found when googling ‘linux desktop market share’.
http://www.zdnet.co.uk/insight/software/linuxunix/0,39020472,391186…
2.8% two years ago.
Sorry everyone, but it was probably Linux users like myself skewing the results. I use IE through wine to get flash working on my 64bit Linux machine
I think it’s funny that’s the easiest solution I’ve found so far
Cripes man, why not just install a 32-bit firefox? I’m impressed you got that to work; it can’t have been easy.
No site done in flash is worth that much effort!
4. Macintosh 2.32%
8. Macintosh Power PC 0.15%
As far as i understand, there are 15 times more intel mac users than powerpc mac users????
^_^
Silly, they’re talking about Classic.
Well then they should say “Classic Mac OS”, not Mac Power PC. I’m running OS X 10.4.7, and I’m also running a Mac with a Power PC processor. I think this is even worse actually. Now it isn’t just a survey error, it’s a matter of the survey makers being careless and having a lack of knowledge.
“As far as i understand, there are 15 times more intel mac users than powerpc mac users”
According to Wikipedia, MAC_PPC was the useragent string in IE up until IE4.
http://en.wikipedia.org/wiki/User_agent
Onestat (from another site) still registers some IE on Mac users as “Macintosh Power PC”.
The latest distros, like Ubuntu or OpenSUSE or kanotix etc are still trying Hard to get efficiency of Win2000 in terms of EASE OF USE for average users.
When will they compete with XP and in turn Vista? maybe in 2012…
If linux beats Mac It is more than enough to celebrate and drink 12 pk beer.
http://www.thecounter.com/stats/2006/August/os.php
And you wonder why I always say Linux is 1% … I was being generous.
The results are obvious. Linux users tend to do programming and major server work. Windows users tend to type numbers into excel spreadsheets and play solitaire. The linux users are just too busy with real work to spend all day browsing websites!
The results are obvious. Linux users tend to do programming and major server work. Windows users tend to type numbers into excel spreadsheets and play solitaire. The linux users are just too busy with real work to spend all day browsing websites!
Linux users busy with compiling Linux kernel; they try every possible options only to gain 0.01% performance up. They also busy to try tens of mp3/video players to find which one is the best emulating WMP. They also busy to send complaint emails to every hardware vendor who didn’t open source their drivers. Also most of them only play KMahjong ang Tux-race all day.
Speak for yourself. After am done installing my new distribution for the week, getting all of my plugins and codices working, and reading these blogs I don’t actually spend that much time working. Fortunately, when I do work I don’t worry about viruses or popups and I get right to it. These days I running OSX on my macbook and I’m a bit grumpy about not having as much fun as I used to have. Hopefully Redhat will get their stuff together and make something that I can use.
After more than 10 years struggling on the desktop, Linux has a bare 0.3% market share. I think it’s time to give up and to install a Windows XP + good antivirus on all your computers instead.
Thanks, and will you be paying for all of that?
I’ll be glad to tell you the address where to send the money to buy licenses for my home-built pc’s. And laptops. And servers.
So stfu and use windows if you want. I’ll use what I want. I happy, and you, I don’t really care what you do.
The whole point is: Use what works best for you, that doesn’t matter if it doesn’t index high in stats.
The story of Linux and the free software movement in general is that nobody ever thought it would amount to anything, but we did it anyway. Even if we never reach 10% of the desktop market, we’ll do it anyway.
We’re not motivated by marketshare, we’re motivated by progress. From my perspective, Linux and the free software stack is coming along quite nicely, and the pieces are really starting to fit into place better than I had expected. In the time it’s taken Microsoft to develop Vista, a system that (very) generously represents a doubling of “utility” with respect to XP, the Linux distributions have undergone massive improvement that might very well be unprecedented in the history of software development.
Admittedly, there was a lot of low-hanging fruit. However, there’s more low-hanging fruit today than there was 5 years ago, and there will be even more next year. Most OSS projects get more bug reports than they can handle, and the more people get involved in these projects, the more bugs we find. This isn’t a sign of bad code, it’s a sign of progress. As I said, we’re motivated by progress.
We’ve taken a kernel that was designed for a single 386 and turned it into a kernel that scales efficiently from ARM-based cellphones to 4096-CPU NUMA servers. And you know what? It isn’t really that tricky or complicated. With an intermediate background in C, you can skim through kernel/sched.c and see exactly how we implement constant-time scheduling and efficient load balancing. We do it because we can, not because we have customer demand.
Why do we insist on creating such a capable operating system, even though it isn’t clear that we have a significant market? Because if we didn’t do it, then nobody else would*. That would be a shame, because, as we’ve shown, it is possible. You just have to throw capitalism out the window and start hacking.
*Note: I don’t want to take anything away from the many other community OSS kernels and operating systems out there, some of which have rivaled and surpassed Linux in many respects. These projects won’t give up either, despite even more modest marketshare, and I have great respect for them.
Well said.
G.
The free and open source people have also created a new metric for software users. The thing is that before GNU started creating very useful utilities and giving them away very people knew anything about “source” code. Now we have international and national discussions about these things. Although many Linux hackers have turned their back on the historic software industry and are not concerned with any usefulness or practicality associated with non-free software, they have indirectly improved the closed source software by helping to keep the big proprietary software companies honest.
I didn’t think that I would end up having to care about freedom to do whatever I wanted with software when I started my Computer Science degree. After years of frustration with Windows I realized that the situation was only going to get worse and that big companies like Microsoft, Autodesk, Adobe, etc were only going to intensify their hold on me. I realized that they were only going to spend an increasing amount of time coming up with new clever ways to prevent me from doing things that I want to do. I realized that these companies as a whole were more interested in locking me into a perpetual upgrade cycle then providing me with valuable services.
Lets face it. Today computers are POWER! There must be people in the community that know how they work and can change them and verify that they are working correctly. Without this we are all lost. Fortunately, we have it though, so its all good. Peace
That was pretty funny. ;-p
Well 75% of the people know that over 50% of statistics are off 15% and most can skew the other 85% to prove 90% of just about anything 51% of the time. Sampling is not an accurate measure, just somebody’s best guess of an estimate taken from a tiny little sliver about the rest of the pie.
Mindshare now that’s important. How do they feel about those servers?
“Well 75% of the people know that over 50% of statistics are off 15% and most can skew the other 85% to prove 90% of just about anything 51% of the time. Sampling is not an accurate measure, just somebody’s best guess of an estimate taken from a tiny little sliver about the rest of the pie.”
Amen to that. I went through graduate-level statistics. The results of the study could also be interpreted as:
That 96.97 percent of users hit a specific website which, for whatedver reason, attracted Windows users.
Aside from the margin of error there could be other factors which skew the curve. Only 2.47% of Mac users .36% were Linux users were intereseted in the site.
Maybe Mac and Linux users had better things to do. Who knows?
BTW, 96.97 percent of Windows users drink milk. Brought to you via a double-bilnd study by the Milk Producers Association.
At my shop there are 11 Computers.
1 Runs Server 2003
1 Runs XP
1 Runs Solaris 9
1 Runs Tru64 Unix or OpenVMS depending upon the job in hand.
The remaining SEVEN use Linux. Of these, One I use for browsing although, I do sometimes use the XP system.
The other SIX?
Well, the only time they go out to the internet is when I do periodic package updates. They have more important jobs to do.
Like some previous posters, There a many many systems working away thet would never come within range of the sampling done in this survey.
I make this report akin to someone standing by the side of a road and counting the makes of cars driving along. So what if after 10 mins, not one Merc has come by. Does this mean that 0% of cars on the road are Mercs? Nope. Especially if you are counting the cars outside the BMW Factory!
The value of any survey depends upon the sample size and methodology for choosing the samples. Two million users is a fairly large number of samples. No problem there. But it’s not clear whether the samples were culled from site(s) that are frequented by a sufficiently diverse set of clients (ie. Windows, Mac, Linux, etc). For example, you’re going to get very different results for a site that advocates the use of a particular OS (ie. MacFanatic.com, WindowsInsider.com, LinuxWorld.com) than you will from a general-use site (ie. Amazon.com, Yahoo.com, Google.com). Regardless, it’s an interesting anecdotal survey. I don’t doubt the results. Much. But it would be useful to see more data on the methodology.
Here in Poland a the same probe there are 1.1% of foreign (not Polish) Linux users:
http://ranking.pl/index.php?page=Ranks:RanksPage&stat=1|FG
At the probe of ca 10.000.000 Polish users there is 0.8% of linux users:
http://ranking.pl/index.php?page=Ranks:RanksPage&stat=1|OW
My computer often lies to the internet. When using Oprea My Suse base computer tells the world that it is IE on Windows XP – Why — because there are sites out there that crap out when you tell them anything else. Sooo if the sites just ask politly – they are missing nearly all Opera trafic. While Opera is only a small (but unknow) trafic it is just another way the number are inaccurate.
PS. I know many of my clients end up using Opera as it has more functionality than FireFox out of the “box” and the computer newbies are often unwilling to modify their software. We only give out systems with Linux on them.
Safari User Agent Strings
Change it:
http://www.celestialfrontiers.com/safari_enhancer.php
Analyze it:
http://www.useragentstring.com/
For those who don’t believe the Linux and Mac numbers then take a look at the W3schools OS statistics:
Mac: 3.6
Linux: 3.5
You’d expect that in a relatively technical site more people who’d be using Mac or linux… Well, I guess this put an axe on your doubts.
If tech users don’t care about linux/mac then these results are even more obvious.
–
Did OneStat monitor:
a) microsoft.com ? and
b) Windows Update ?
If I use a Mac OS – would I frequently visit microsoft.com or windows.update.com – nob!
As a Mac user, would I visit my country’s private and national radio & tv sites (mostly for Windows only), – no I will not! I may subscribe to RSS feeds or Podcasts from them but not visist their Windows_Media_only sites more than once. Would I visit apple.com/software/, yes I would.
Would Linux users visit apple.com/quicktime/ – no, they would not ! Would they visit CNN.com, not as much as Windows users. Would they visit microsoft.com – no ! certainly not in any quantity ! Would they visit slashdot or osnews, yes!
Best wishes
Pravda Statistical
Ask the ISP’s
.
It has been reported that, for the most, patients do visit their doctors.
.
Even if the share reported is accurate for all systems, several million Linux users form a very nice self-sustaining market.
If we say there are a billion internet users as someone suggested, 3600000 desktop web surfing users is a very nice foundation. I think the larger issue is that those 3.6M are split among numerous distros.
Apple’s share which is around 25 million is also very nice. Those who say they should just give up because they lose market share shouldn’t ignore that. Many folks in many industries would be happy to have 25 million captive, locked in customers. Sure, there are incompatibilities with the other 969M +/- users, but so what? 25 million is 25 million. That can sustain itself.
First and foremost, these figures means little, if any.
Out of the 11 machines I have (both @home and @work), 2 are Windows XP and 9 are Linux. (I write cross platform drivers/software).
Out these 11 machines, only 5 have ‘net access:
Work: Development workstation [Linux]
Work: Network testing [XP]
Home: Development workstation [Linux].
Home: Laptop [Linux]
Home GF’s computer [XP {and BSD}])
If someone takes stats, I’m 50% Windows and 50% Linux.
More-ever, while I usually browse technical Linux/Computer sites, my GF’s usually visits every-day sites so in essence, if anyone meters my (ours) browsing habits targeting ‘normal’ sites, he’ll most likely come to a conclusion that 90% of our web traffic is Windows XP and 10% is Linux, while in reality, it’s ~82% Linux and ~18% Windows.
The title of this piece should have been: “97% of the desktop machines that were used to browse the following list of sites were using Windows”.
Now to my question:
I can understand that MAC/Linux/BSD is not for everybody; Heck, I write Windows code on an (almost) daily basis and I spent hard earned cash on an XP license for my GF.
What I -don’t- get is: Why are you so happy to read these stats? You don’t like Linux, OK, Good for you, but why are you so happy?
The only reason you have a chance in hell in getting semi-working-almost-bug-free-no-as-insecure-as-XP-Vista is due to MAC and Linux breathing down Microsoft’s neck.
Without Linux/MAC/BSD, you’d be paying 1000$ for a single year Windows ME 2007 Live! subscription and pray to the God all-mighty that Bill won’t decide to double the price next year.
If I was an avid Windows user, I’d be happy to see Linux and MAC gain a 50% market share, forcing MS to spend -more- money on R&D, more money on actual innovation (as opposed to adding tabs to IE7, 5 years after everyone else) and less on stupid !IsNot patents.
I just don’t get it.
(EDIT: Spelling… me -bit sleepy)
Edited 2006-08-14 21:23
http://www.thecounter.com/stats/2006/August/os.php
and
http://andrewhitchcock.org/companystats/
“Methodology: A global usage share of xx percent for OS Y means that xx percent of the visitors of Internet users arrived at sites that are using one of OneStat.com’s services by using the particular number of OS Y.”
Which basically means you have to visit one of their sites to get counted. OneStat have pretty big delusions of grandeur if they think they’re the whole web.
Google used to do OS stats.
Last batch (June 2004) here:
http://www.google.com/press/zeitgeist/zeitgeist-jun04.html
XP – 51%
Win2K – 18%
Win98 – 16%
…
Linux – 1%
Nobody uses Linux. They didn’t in 2004. And they stll don’t.
Of course someone will argue that Linux users didn’t use Google in 2004. And 95% of us will laugh at them.
> Nobody uses Linux. They didn’t in 2004.
> And they stll don’t.
Gee… guess what I’m using to reply to you. Paint with large brushes much?
(Mozilla/5.0 (compatible; Konqueror/3.5; Linux 2.6.17-gentoo-r5) KHTML/3.5.4 (like Gecko))
Edited 2006-08-14 23:35
“Paint with large brushes much?”
To counter all the hype over an OS with .4% of the market?
Sure.
.4% of the web browsing market, and what, 35% of the web serving market?
Yes, hype, it’s just hype. No one with any brains is using this fad.
http://www.nsa.gov/selinux/
One out of every 100, yea, that’s nobody.
Why if I had a dollar for every 1 out of 100 people I’d have … 60 million dollars
.
“… Linux 0.36%”
“Please note that OneStat bases its numbers on… extraction from user agents.”
Then these results are legitimate. Because no self-respecting linux admin or user would allow an “exctraction agent” to run on their system.
Of course MS has a higher market share, but there is no way that Linux’s 0.36% is legit.
Um, every web browser sends a user-agent string with each http connection. Even wget sends one! And it virtually always includes the OS and browser. Some let you send false ones (like wget and Opera) but yes virtually every linux user voluntarily gives up: OS, arch, browser with each http connection…
Wow.
Oh… pls forgive my misunderstanding. I wasn’t aware that this is what they ment when they said “extraction agents”… I was thinking more along the lines of software that sends information to them (spyware or not)… anywas… I guess my comment no longer stands.
I do think that the websites may have been biased (I mean, how many linux users browse http://www.windowsitpro.com?) but yeh. These results are hardly trustworthy.
“These results are hardly trustworthy.”
Yeah. Google was lying too. They all do!!!
(Sarcasm blatant)
That’s definitely true. That’s why I think Google should be able to get the best accuracy. But they’ve apparently stopped giving out their statistics, probably sick of watching people quibble over them.
“extraction from user agents”
user agent is a technical term that means web browser. like IE.
thus extraction from user agent means, collecting information sent by browsers.
“extraction from user agents”
it certainly DOES NOT mean: an agent of extraction from user.
Well, that’s kinda depressing.
I guess, depending on where you look, the number is somewhere below 5% (I mean reasonable numbers), you just have to choose which number exactly to believe.
Here are some more stats, with similar very sad results for Linux:
http://www.thecounter.com/stats/2006/August/os.php
Windows: ~94%
Mac: 3%
Linux: < 1% (and that’s *well* under 1%, it’s actually rounded down to 0%)
http://marketshare.hitslink.com/report.aspx?qprid=2
Windows: ~96%
Mac: 3.80%
Linux: 0.44%
Yep, after nearly 10 years of nearly uncritical praise in the mainstream press; after 10 straight years of getting as much, if not more attention in the tech media than Windows; and after 10 years of being free (as in beer), Linux has ~0.4% share.
You do realize that the largest amount of press around Linux has been more interested in the bazaar roots rather than ending Microsoft.
Ending Microsoft is a buzz story you use to get more hits, the serious discussion has been the impact on how we develop software.
If you’re so 2-sided on life that you can’t see these stories for what they are, I’m sorry. But you really should be able to tell at this point they’re made up of two things (and I mean the “Year of the Linux Desktop” stories):
1.) Geeks who are excited about more people liking the same things they like.
2.) Reporters looking to score easy hits for their news site.
The first one you just can’t despise: If you don’t like people who want others to like what they like you need to just stop socializing. Finding a real problem with “hey, I like this, I bet you would too” is pretty difficult, and you’ll have to misconstrue it as forced evangelism to not look like a jerk (hence the term Zealout). And the second one, well, reporters are the minority in the world.
If you want to know why Linux is so neat you need to look at user groups filled with free help and local meetings. Free installfests, just to help others enjoy the same things they enjoy! Software developed in a way that probably made Brookes gasp. And it’s certainly not just Linux, but Linux is certainly the “poster-boy.”
P.S.- The emotion I see in your posts indicates anger. It’s unlikely that anger proceeds from a satisfaction of desires, and given your post I doubt you have a problem with Linux failing. So, tell us, what has you so emotional? Do these results surprise you? .37% on onestat doesn’t surprise me!
and after 10 years of …, Linux has ~0.4% share
So what did you expect?
Linux is designed by geeks for geeks.
Linux designed to be USABLE will lose all of it’s power.
Anyway all modern OS-ses are crap.
Linux designed to be USABLE will lose all of it’s power.
C’mon, that’s simply not true. Sure, preventing power users from gaining access to a command-line is going to hinder then — but that’s not necessary in order to make Linux more usable.
Here are some more stats, with similar very sad results for Linux:
http://www.thecounter.com/stats/2006/August/os.php
Windows: ~94%
Mac: 3%
Linux: < 1% (and that’s *well* under 1%, it’s actually rounded down to 0%)
Yeah, this is an excellent example site to show how people and their statistics can lie to you pretty bad.
Because, you see, this site has global stats (including OS stats) going back to 2001. What we see ?
A surge of Linux users attaining 1,200,000+ in 2003, and then, it goes dramatically back to 730,000- in 2004.
What exactly happened with this site and Linux users ? Because Linux users numbers always increased till today,
but this site lost nearly half of them in one year (perhaps less, you can check month by month, I just checked march
stats for every year). So the stats of Linux users is just useless, but they sure are used by some people to say things
like “Yep, after nearly 10 years of nearly uncritical praise in the mainstream press … Linux has ~0.4% share”.
When you read that, when people state that these stats actually represent Linux users as a whole, you know they have
an agenda, rather than searching the truth.
Monopolist controls web OS market…
Seriously is this a surprise to anyone? Or even newsworthy?
Would be interesting to know what proportion of that 96.97% had “HotBar” in their user agent somewhere 🙂
Incidentally, what are the ninth and tenth most popular OS’s? They seem to have a bit of a blind spot about the difference between 8 and 10 😛
* cough * * cough *
Less than .5 percent? That’s so sad. Even with the all out media blitz (osnews and others) to promote Linux over the last 6 years. This is a testement to the failure that is the Linux OS.
I don’t care if linux is a “failure” or not. After years of viruses, worms, spyware, etc. and hours of maintaining my Windows system (and the cost of firewalls, anti-virus, etc.) I moved to linux. Now I spend my time using my system, not maintaining it. Do I care what the numbers say? Not one bit.
>>Do I care what the numbers say? Not one bit.<<
Then I guess you don’t care about continued HW/SW support either. HW/SW makers don’t care if their products don’t work for 0.4% of the market.
>>Do I care what the numbers say? Not one bit.<<
Then I guess you don’t care about continued HW/SW support either. HW/SW makers don’t care if their products don’t work for 0.4% of the market.
If linux really were at 0.4% do you think it would have the HW/SW support that is available today? Obviously someone thinks it is worth doing regardless of what OneStat claims.
I use Linux frequently, and the fact of the matter is that there are plenty of devices that aren’t supported. One of the worst areas of support is wireless adapters. I can’t tell you the number of times that I’ve had to hunt around for drivers, not find them, and have to try to hack together a solution — the number of hours that I’ve spent is ridiculous. So, clearly, there are manufacturers that care about marketing statistics. That affects everybody.
Support for server hardware is usually pretty good, but support for customer hardware is lacking behind. It’s not really surprising, given that most companies backing Linux are interested in enterprise solutions.
That said, the slow Linux penetration in the desktop market isn’t that surprising. MS-Windows, even with all its flaws, does the job. The overall penetration ought to be higher than 0.4%, though, while taking consideration of the server market.
Then again, these numbers don’t matter much to me. Thanks to Linux, I have all the functionalities I need without having to spend money on CAL/licencing subscriptions or resorting to piracy.
The methodology used in this press release isn’t clear. Were they counting hits or visits? If it’s the latter, then what is a visit for them?
After reading the reader comments above, it is clear that Linux users don’t care if Linux has only .5 percent of the market. Ah! But they should. Here’s what puzzles me:
How can the Linux community not care about an accurate statistic like this, but be so deeply passionate about Microsoft getting sued. Is it because you want the government to do what the Linux community has failed to do? That is, take down Microsoft?
Now I read comments like, “use whatever you want, who cares…” I thought it was about freedom from the Bill Gates Tax? Looks like that arguement died a long time ago.
All this really takes me back to the mid-90’s when Linux/Unix users took pride in the command line environment and vowed never to give in to a fisher-price-like “clicky-pointy” user interface.
It seems to me that all the chest beating (and knuckle dragging) going on the Linux community is really just an effort to hide the fact that their operating system not only sucks but is also a complete failure. (but you wouldn’t think that given all the media attention).
The media blitz has failed.
I think it is time the linux community stop being a bunch of mavericks and they should start listening to the Windows community. The Windows folks obviously have experience building an operating system for the people by the people (YES! Microsoft REALLY DOES listen to it’s customers).
posted this one in the wrong thread, sorry.
Edited 2006-08-15 05:41
Than you’re just not well informed. As a matter of fact, Linus Torvalds spoke several times about Linux world domination.
Also see this article from Eric Raymond.
http://www.linuxjournal.com/article/3676
Not that this would be a bad thing…
Edited 2006-08-15 05:42
Nice 🙂
I just completed (in six weeks) a complete migration from XP to Ubuntu and feel safer and happier with a very strong and friendly community support and a top-notch quality and easy distro. After three or four short different tries for a Linux distro in the last three years, I can say now that I really switched for good and miss nothing.
So, I can’t recoup these statistics with my experience and just smiles.
What to say more? I wish you all the best 😉
I think it’s depend what they were checking.
for example my homepage is not place for any OS users. It’s just homepage.
So my visitors are 60% Windows XP users, 10% Linux users, 8% xBSD users, 4% windows 2k users, 2% win98 & about 1,6% MacOS users. it’s still not so many of the Linux users there.
I maintain an Au Pair agency website. This is a totally non technical audience, spanning most of the globe. The stats are as follows:
Total unique visitors: 348’608
Last 31 days: 28’219
84.3%: Internet Explorer (inc. Maxthon & AOL)
v5* (288 views) v5.5 (102 views) v6 (21’553 views)
12.8%: Mozilla Firefox (inc. Netscape & SeaMonkey)
2.1%: Apple Safari (inc. Linux Konqueror)
0.8%: Opera Browser*
OS Breakdown:
(This Month)
94.3% Microsoft Windows
9x (873 views) NT (84 views) 2K (2258 views) XP (23407 views)
2.8% Apple OSX
0.4% Linux
I think these are pretty indicitive of what people have been discussing here.
The article states “The OneStat.com solutions provide executives, marketers and webmasters with answers to critical e-business questions”. I would like to know who paid for the survey. Of course, OneStat would never skew the survey to report what their customer wanted the results to say, even when they are selling to Marketers.
Sadly, I don’t think OneStat is a msft shill company. Consider this from a recent OneStat survey (it’s not exactly msft praising):
>>
Firefox Usage Passes 15 Percent in US
By Nate Mook, BetaNews
July 10, 2006, 2:39 PM
Mozilla’s Firefox browser continues to post gains in market share, according to Web analytics firm OneStat.com, while usage of Internet Explorer has fallen more than 2 percent since May
<<
http://www.betanews.com/article/Firefox_Usage_Passes_15_Percent_in_…
I am a linux/KDE/Konqueror User… I have to say that I only visit a small number of sites regularly… Most of these are Linux/OSS related or CG Related with the exceptions being one or two dating sites (mostly okcupid if you must know), ebay and last.fm I think you will find that a lot of linux users have simular interests… and so will only show up in significant numbers under certain types of sites.
Having said that the number is probably still quite low. But it is large enough to be self sustaining…
As a Linux, BSD, MacOS, and *nix user I could really care less about MicroDump. It’s a side effect if it eventually lands in the trash heap of history as the most expensive corporate joke on consumers.
Oh. BTW. One of our WinDump 2003 servers took a dive on Friday. Woo hoo! Feel the POWER! Not.
Edited 2006-08-15 12:55
Is that we are millions of linux users ! and most of us also use Windows every now and again meaning that we are even more than suggested.
I read logs and reports daily for a pretty large website plus my own hobby domains, and these stats seem about right to me. However much people would like to dismiss these results (the stats are a bit ugly), I see no other evidence to suggest that they are inaccurate.
To me, this says a few things:
(1) Linux has done well on the desktop in terms of the overall quality of the experience in spite of our numbers. I think the fact that the number of users who use Linux is so low compared to Windows says a lot of good things about how much a small community can do.
(2) Linux could have a bigger percentage of the desktop market if it had more advertising behind it targeted toward consumers. Fact is, there’s almost none. At least not on TV. Then again, think about the kind of nasty commercials you’d need – you’d need to promote a “Linux lifestyle,” the same way Apple has done. Bleargh and all that; I’m glad the tenor of the Linux community is what it is, which is tolerant of commercial ventures, and some flourish, but is overall anti-establishment and most things are free as in beer as well.
Personally speaking I hope that never changes. I think it’s worked well. I *like* my desktop. I’m really not overly concerned about what the world thinks except when they do things like lock certain browsers out of their site. But for me, this has been rare. Linux is ready for *my* desktop, if not for others.
(3) Hardware vendor support, which someone mentioned, is a valid major drawback. There’s nothing positive about this. When possible, we must all be in the habit of researching Linux compatibility/support first, and only then buying hardware. I’m sure most Linux users are.
(4) All of those using Linux really *are*, when all is said and done *that unbelievably l33t*. I say we make like Holden Caulfield and go watch ourselves get tuff in the mirror. We are the < 1 percenters. We are dawn riders at the gates of oblivion with huge massive phallic lances and swords and stuff riding fast at the gates of a fortress in Redmond which does not notice us. Maybe they never will but WE LOOK COOL AS HELL, THOSE OF US NOT TOO OVERWEIGHT TO CRUSH OUR HORSES. If anyone gets like, chicks, because of this, let me know.
(5) Perhaps relevant, perhaps not, are the fascinating weekly stats posted on the otherwise lamentable comp.os.linux.advocacy newsgroup, a nasty place that I do not recommend, but which indicates that many Linux die-hards post their messages from Outlook/Outlook Express. This is a source of perpetual mockery and for the most part, rightly so, especially when people called on it do the equivalent of text stammering to explain themselves. Well, this used to be the case; I haven’t looked in a year or more, so hopefully this has changed. Making a clean break with Windows for many people is a hard thing. Heck, 5 years later I still have a Windows box though now it is used only for video editing (and that may change when I get firewire in my new Linux PC).
I still don’t think Linux is for everyone. I think it is for more people than currently use it (there are a lot of sharp people out there into computers who really have no reason to be running Windows, at home especially), but things which were once true about Linux (hard to install, hard to use, “not ready for the desktop”, etc.) dominate many discussions as also being true *today*.
Well I just installed OpenSUSE, and I cannot see where that is any harder to install or use than Windows, and I’m trying to be objective. Gentoo may not be for the masses but OpenSUSE and probably Mandriva (haven’t used it in awhile) and various other Linux distros are.
A few days ago, there was an article here on osnews.com about Ubuntu – I think no matter what distro you use, it’s worth getting behind Ubuntu as an entry point for those considering switching from Windows. It would be nice to have one distribution that is the main “gateway” into Linux, with appropriate built in system help and application naming schemes and so on to facilitate this change. A distribution specifically geared to switchers, with an attendant community focused on the “Windows way of thinking” and supporting users who are “programmed” to think Microsoft (things like drive letters and GUI first, command line running on GUI rather than reverse in Linux). Once people switch, they can then explore alternatives. Whatever distro this would be (And I think Ubuntu or Kubuntu are great possibilities, but that’s just my opinion), it would be a pretty good support sink for new Linux users. I don’t think anything really fills this need (and no, Lindows or whatever they’re calling it now, is not a positive alternative IMHO. It should be a free as in beer distro at minimum).
It’s a hard thing to convince someone to change their OS. It is *disruptive*. Many of us enjoy this disruption because we enjoy new things and exploring and tinkering. Most people loathe it. This is the biggest obstacle. Then, add in the lack of any advertising about Linux – sure we on osnews.com know all about it even if we don’t use it, but does your mom? Probably has heard of it but doesn’t know anything about it. But she knows what a Mac is because of the fairly inventive commercials…if she wants an alternative, what’s going to pop into her head?
What we should do is pool together contributions to get some good TV spots going – that is, if spreading Linux is an important goal (I don’t know if it is). Focus on a lot of screenshots/what the desktop looks like. Show all of this cool stuff you can do with Linux and end the spot with, “Oh, and it’s free.”
Alternately we can just be happy with what we have now – barring lack of vendor hardware support, which as I say, there’s no positive way to look at it, I think things are going fairly swimmingly. Of all of the things I’d like to see change about Linux, getting more people using it isn’t even in my top 5. It may be at the bottom part of my top 10.
I read logs and reports daily for a pretty large website plus my own hobby domains, and these stats seem about right to me
Nice introduction to make people believe you are an authority on the matter, classic FUD tactic.
However much people would like to dismiss these results (the stats are a bit ugly), I see no other evidence to suggest that they are inaccurate
Sophism to add to the FUD tactic (“I see no other evidence”, coupled with earlier “and I’m an authority on the matter”).
And no, nobody dismiss these results, people say it’s not accurate to take those for finding the true Linux numbers accessing web pages on their desktop.
It wouldn’t be any better if taken on a Linux dedicated site (where you’ll see lots of Windows users, being Linux users accessing from work).
To me, this says a few things
Which is complete BS, as you don’t even know the methodology or sites checked.
Linux has done well on the desktop in terms of the overall quality of the experience in spite of our numbers. I think the fact that the number of users who use Linux is so low compared to Windows says a lot of good things about how much a small community can do
BS damned praise ! Linux started from far lower numbers to attain the state it is in now. No need for a stupid Web survey to acknowledge what you talk about,
we knew that already !
Linux could have a bigger percentage of the desktop market if it had more advertising behind it targeted toward consumers
What a prophet ! Let me make my own prophecy : Linux could have a bigger percentage of the desktop market if it came preinstalled on far more desktop systems.
BTW, we don’t need market share for market share, this would be stupid. You have to support all these people, but not warez oriented people.
All of those using Linux really *are*, when all is said and done *that unbelievably l33t*
Yeah, like my computer illiterate wife or kids, who don’t even understand what l33t means.
We are the < 1 percenters
No we’re not ! What is your metric BTW ? 1 percent of what ? You don’t even know. I’m not from the USA, and I can tell you that USA must be one of the worst country as to percentage of Linux users. You’re a better astroturfer though, but are revealed by stupid crap sentences ending with “Redmond not noticing us”, which is BS everyone can see. I’m not even talking about the insults contained in your sentences, about Linux users.
Perhaps relevant, perhaps not, are the fascinating weekly stats posted on the otherwise lamentable comp.os.linux.advocacy newsgroup, a nasty place that I do not recommend, but which indicates that many Linux die-hards post their messages from Outlook/Outlook Express
And it never occurred to you they did that from work with a Windows only policy ? Instead you rely on perpetual mockery (I know you talked about yourself).
These stats are just that, statistics. Without even the sites checked, you can’t guess how accurate it would be of Linux users on the desktop surfing on the Web.
Like that matters anyway. Actually, this survey is irrelevant. What we want is support, so stats on NVidia drivers, ATI, Pinnacle, Hauppauge, … web sites would be far more useful (though still not accurate) than this stupid survey.
As (probably) a majority of Linux users dual boot, the stats may in fact be a bit off. I’m not suggesting that they are wildly inaccurate, but think about it this way: I’m currently at work, using IE6 on XP, but on my laptop, I use nothing but Ubuntu. So, if I hit the same pages which this survey considers, then I show up as 1 XP user and 1 Linux user. In my case, the survey may accurately assess hits or visits, but does not accurately assess the users. This computer is a work PC and has no exclusive user. My laptop does and it is all Linux. Yet I still will show up on the survey as two users.
If 2/3 of Linux users dual boot, and I think that is probably on the low side, and they spend half their time in each, then fully 1/3 of Linux users would show up twice, once under Linux and once under another OS.
There is NO WAY microsoft has a 97% share. My personal website hit stats show windows at 50%, linux at 38%, apple at 11%. Now, granted that I do not represent what is out in most of the world, but additionally judging from the people I encouter, polls I’ve done on various forums, etc … microsoft has nowhere near that percentage, not even close. My guess is that they’re closer to 80%. I’ve also seen stats from large websites and those are never anywhere near 97%. And another stat that’s just plain wrong is apple intel vs ppc – no way intel has such a 15X marketshare over ppc, there’s just too many ppc macs out there.
This study is yet another “independent” think tank that took money from MS and came up with the numbers MS was looking for. Also, since onestat.com deals with corporate websites, naturally that’s going to skew in favor of windows, even if they weren’t paid off by MS.
Those are interesting opinions — but do you have any actual proof that this survey was funded by MS?
“And another stat that’s just plain wrong is apple intel vs ppc – no way intel has such a 15X marketshare over ppc”
Onestat doesn’t say that.
What they did say was that .15% of visitors were Macintosh Power PC. Another group were just identified as Macintosh. There was no stat for Intel Macintosh.
Linux has a trivial market share. Those are the facts.