Submitted by It’s market share/installed base/whatever time! Net Applications’ figures had been under review for a while, and now we know why: they’ve finally done what should have been done ages ago. Namely, they’ve added country-level weighting to their figures. They also added this retrospectively, meaning all previous figures have been updated to reflect the change. This has devastating effects for some.
Country-level weighting is quite important when it comes to the way Net Applications collects its figures. “We adjust our reports proportionally based on how much traffic we record from a country vs. how many internet users that country has,” they explain on their website, “For example, although we have significant data from China, it is relatively small compared to the number of internet users in China. Therefore, we now weight Chinese traffic proportionally higher in our global reports.” This makes the worldwide view of these statistics more accurate. They use CIA data as the source for internet users per country.
So, what does this mean? Well, for operating systems, since the Mac is much more popular inside the United States than outside, this change has drastically reduced the figures for Mac OS X: worldwide share “dropped” from 10% to 5%. Windows went up from 88% to 93%, while Linux remained stable at its 1%.
The browser view per version shows a much more grim view: Internet Explorer 6 is now, once again, the world’s most widely used browser (according to these statistics, of course), holding a 27% share. Internet Explorer as a whole gained a bit at the expense of Safari, while Firefox remained stable at 22%. Opera’s share goes up to 2%, reflecting its popularity in Eastern Europe and Asia. Chrome has a share of 2.6%.
Happy commenting.
Things look much more fair now, specially OS X going down to 5% from 10%. However, I find it very sad to see IE 6 as the most used browser world wide. I was hoping that it would die this year, especially since Youtube announced it will stop supporting it soon.
It would be more interesting to see if youtube really DOES stop supporting it based on these figures.
You know, I still don’t buy that 1% figure for Linux. With all the tech news talking about massive rollouts in the tens or hundreds of thousands of computers in large companies or governments, the figure must be closer to 4%, at least.
The squeaky wheel doesn’t always get the grease.
Dual boot systems alone put such a low figure in question. Given that the stats are based on web browsers and the OS they report to be run on, I’m not sure how to account for the difference though. I’d love to hear theories on the discrepancy. Intuitively, it doesn’t seem right but I can’t figure the math behind the missing link.
As for IE6; why won’t someone just take it behind the barn and shoot it in the face. That lame duck has been limping along far past it’s time. Come on big business, you’ve had years to upgrade your in-house webapps; why do they still rely on IE6 corruptions of the html standard?
Question is, how they measure OS use. E.g. if they get the info from browser data, then only desktop machines with active browser users will show in the statistics. Also, a lot of companies have thousands of Win desktops with a lot of users who surf a lot during the day. I don’t know how they build their statistics, but I’d say there’s still no reliable one.
I don’t think computers in large companies or government networks would show as different machines on web servers’ usage stats.
They all would share the same IP and probably show the same OS and browser versions.
That certainly could account for the discrepancy. And for that matter, what about homes with more than one computer? My brother runs Windows religiously (though right now he’s learning some Penguin love because he’s forced to use a laptop that isn’t his
), and I run Linux. How would Netapplications deal with that?
FWIW, during some recent data protection training we asked all 120 staff in our company what operating system they use at home. Three (all senior managers and none from the IT department) use Linux and everyone else uses some Windows variant. No-one uses OS X. If we assume 50% of web surfing takes place in the office, our small and unrepresentative 1% figure for Linux seems about right. That data is based on financial services workers in a provincial British city.
Our corporate web site stats tell a broadly similar story – 1% Linux, 2% OSX, 1% mobile devices and the balance Windows. Our customers are typically middle-earning, middle aged British heads of household.
In summary, the restatement of Apple’s share to 4% certainly ties in better with my own experience, and Linux feels about right.
My question really is this .. how much is the margin of error that the “C.I.A. data as the source of the number of internet users per country”? I mean things link how many internet users that country has can be very circumstantial. For instance .. are there country’s where based on whatever the metric for the user may be is biased, as one person has multiple computers (at home, at work,etc) or multiple people share usage of on computer (classrooms, internet cafe, library access)?
I always look at number with a careful eye as they can hold such deep meaning.
so we can see who is using what in which country.
All this ‘weighting’ nonsense sounds really fishy to me.
The world is not the USA.
The world still a big multicultural planet with many countries following their own way.
Its a nonsense what they are doing with these figures.
Edited 2009-08-03 02:09 UTC
You’re clearly missing what they actually did which is an attempt to improve the accuracy of their data with that weighing.
It goes without saying that country weighting is wide open to misuse and abuse. Just look at NRD data for video games. They’re US-only and they don’t include digital distribution. It was no wonder everyone was looking at it and saying video gaming on the PC was dieing while Valve et. al. were crowing about their digital distribution growth rates.
It’ll go the same way with this latest brain damage from Net Applications. People won’t want to take the time to understand country weighting, so they’ll simply move on to some other set of statistics and use that instead.
This as has been noted probably only a measure of the systems that browse the Internet.
Of the 20+ systems I manage Only TWO ever use HTTP outside the double firewalled LAN they connect to.
One of the two is a MAC and the other Windows. All bar two of the other systems are Linux with the remainder running Windows Server 2003. However, none of them use HTTP outside their private LAN.
I’m sure this is not a unique situation.
AFAIK, these figures are just the sort of ammo that is needed for Balmer in his PR exercise to promote Windows 7 by rubbishing OS/X and Linux saying they are insignificant players in the O/S world.
Why.. oh why are we normalizing in reference to something unrelated??
If you want a market share value, find the size of the overall market and find out what percentage of that market does something. Simple.
NOW, we decide to weigh people differently? Stupid statisticians are always screwing this up!! If you are looking at GLOBAL trends, you look at GLOBAL values.
ONLY when looking at LOCAL/NATIONAL trends do you break down the data into those locales. Normalizing the data into a mean on false pretenses is just bad math! Oh.. wait… statistics is all about ignoring some known data in order to guess at the real value…
So what data is missing? The actual global number of users. Well, take the CIA data, add it all together, then find your share directly. Duh. No reason to break the data down into all these locales and giving each locale a different weight.
All that accomplishes is counting a single China-man’s visit as 2/3 of a visit, and counting an American-man’s visit as 1 1/3 of a visit. Kinda dumb.
–The loon
Well, they probably have statistics from let’s say 10.000 English sites. And statistics from 100 Chinese sites. So if for example, in the English sites 10% of the visitors are using OS X and on the Chinese sites, only 0.1% are using OS X, the final statistics will say that ~10% of the world wide population uses OS X. How accurate is that? If China has, say, 300 million internet users, it is unfair to not give more weight to each of those 100 sites in Chinese than to each of the 10.000 sites in English. I don’t say the math is perfect, but it’s better than it was before for sure.
I should have read the linked article, it explains what they did well enough to actually make sense ( the meaning of country-level weighing just didn’t click right ).
I should make a point to always look deeper when something that should be, relatively, simple seems to be done so wrong…
–The loon
hrm.. now only if I could remember the correct term for the weighing algorithm… oh well.
This statistic IMO truly shows the depth of which pirating is used throughout the poorer sections of the world. My speculation is that if this could show statistics based on Windows version that XP 2000 and 98’s marketshare would be much higher than the harder to pirate and resource intensive Vista. Also this shows why the hold of viruses is so strong in places such as china considering the old unpatched pirated systems used.
Due to server usage i would weight Linux at 2% at least, but for desktop and workstation use (even including dual boot), 1% sounds right to me. Sad thing to me IMO is that many 3rd world computer users would benefit from the better native language support and lack of viruses that a distro like ubuntu offers.
As for Apple, considering the lock they have on their OS (no one in the 3rd world is gonna match hardware to make a hackintosh) I’m surprised at even a 5% marketshare and this really goes to show how much strength they have in America.
China is extremely windows-biased. IE usage: 97% , Windows: over 99%!!
Now, considering that the count of the Chinese population is mind boggling, which company do you think is behind the decision to not only count the users but also the whole population?
A hint, Microsoft is one of Net Application’s “sponsors” (the biggest I suppose) and “by coincidence” Microsoft desperately needs good “stats PR” as their sales and market share is on steady decline (unlike Apple which is also a “sponsor”).
And last but not least, how exactly do they put the population numbers into stats?
If they would monitor only sites for Mac OS X products, the likelihood of Windows visitors are almost negligible, right? Similarly, the opposite would likely be true.
It seems to me it’s like counting processor clock speed. It matters within a processor family but it has little relevance outside of it.
The tug of war is interesting between platforms and browsers but only just. If they start monitoring many more sites with a certain bias suddenly, is there a real increase to a certain platform or is it just within their data?