05.14.09
Is Net Applications One Order of Magnitude Off Target?
Summary: Another go at a comparison between truth and fiction
SURVEYS that encompass many Web sites cost a lot of money/effort/time to produce and the gain from them is mere unless a particular party gains. It is always very, very important to check where the money comes from. It’s valuable to at least understand the business model.
Here at Boycott Novell, 45% of the pages this month are served to “Linux” (that’s more than the equivalent figure for “Windows”, where “Windows” is just the way it’s parsed by AWStats, although there could be HTTP header modification because some people wrestle with stubborn IE-only Web sites).
Moments ago we wrote about the numbers from Net Applications and while we claimed that they were utterly wrong, we wrote nothing about what may be correct.
According to our reader, Chip, Net Applications claimed that GNU/Linux had a market share of 0.3% some years ago (subscription needed). In July of 2007 that’s what they are said to have claimed. But wait. This has got to be wrong because it’s highly inconsistent w.r.t. other figures we have. From the past couple of weeks we get:
Lies, Damn Lies and Linux Market Share Statistics
“The numbers from NetApplications are clearly unrepresentative of reality,” blogger Robert Pogson told LinuxInsider. “Around 2003/4, IDC determined by survey that GNU/Linux was ahead of Mac OS at about 3 percent. Since then GNU/Linux has had growth numbers from 20 to 50 percent in various places.
“That would put GNU/Linux at 7 to 9 percent,” he asserted.
There is also this:
usiness model is rebranding Windows executables and selling them to unsuspecting rubes.
A year ago Matt Assay said it was at 2.02%
ZDNet reported on Feb 24th, 2004 that the 2003 Linux desktop market share hit 3.2% and expected it to hit 6% by 2007. http://blogs.zdnet.com/ITFacts/?p=5334 In 2005 they reported that the 2004 saw the Linux desktop at 4%. I believe that the all the ZDNet figures were spot on. If anything, the Linux desktop market share has continued to increase and is probably currently at 8-10% and rising. Dell and the other PC OEMs wouldn’t have invested in selling Linux pre-installed if it appealed only to less than 1% of the desktop market.
It is quite obvious that NetApplications latest “report” is merely Microsoft’s continuing attempt to control the news about Linux’s success in replacing Windows on the desktop. It’s not working… No one whith half an ounce of brain would take the bait on a “free” Win7 (a dumbed down version that can run only 3 apps at a time) that will deactivate after one year unless the user PAYS Microsoft to activate it. Win7 is NOT free.
As our reader amd-linux (from Germany) put it earlier today:
The thing is that NetApps gets a lot of attention from US/UK IT press, while other observations that show MUCH higher numbers and are equally representative are more or less ignored.
One example is OS stats on heise.de, which is THE mainstream IT new site in Germany. They have now 15% Linux users among their readers, and it would be way more if readers like me were in a situation to use Linux in the office where I read heise.de mostly.
Another example, if you question heise.de being representative is spigel.de which is one of the leading sites of the world (133 in Alexa) and a mainstream news website. Here Linux has around 4%-6%, coming from nearly nothing 2 yrs ago. While spiegel.de is a German news site, is is read from around the world like the BBC site.
It is IMHO safe to assume that Linux worldwide is closer to 2 % than to 1%. The only source who could reliably shed some light on this would be Google, but they refuse to publish any numbers.
But the best, and this nobody can ignore, is the trend. Linux usage is virtually exploding with growth in the double digit percentage area quarter per quarter.
Readers can decide for themselves what the installed base of GNU/Linux really is. We only present some evidence here. █
Gentoo User said,
May 14, 2009 at 4:26 pm
ok, let’s try this again, if only because I’m bored atm.
“Here at Boycott Novell”
Words cannot describe how irrelevant that is, so it’s hard to see why you even bring it up. Wasn’t W3C Schools reporting something like 40% market share for Firefox a few weeks after it was released?
“although there could be HTTP header modification because some people wrestle with stubborn IE-only Web sites”
Ah yes, I remember how the Mozilla market share was “under reported” because of this… until it was no longer necessary because of actual market share and suddenly no one was using it as an excuse anymore. You’re about three years behind the curve.
“we wrote nothing about what may be correct.”
No, because preferences and gut feelings have no bearing on reality.
“(subscription needed)”
Lol, and on the previous post you’re referencing, you whined about the Hitslink data being evil-y hidden behind a paid subscription.
But I take it this data would validate your gut feelings? Funny how there isn’t a single person in the FOSS community willing to fork over $1,000 to settle this? I mean everybody’s writing about how their numbers are wrong, but no one wants to figure out how right they are? Imagine those numbers in Shuttleworth’s hands! Or RedHat! Or IBM! For a measly few thousand dollars!? Of all the people at all the companies that purchase the Hitslink data, not a single FOSS advocate or simply concerned Linux user has ever leaked the fact that the Linux market share in Germany or Thailand is 45%! I think that’s amazing. Don’t you?
“I believe that the all the ZDNet figures were spot on.”
Ha, ha. Of course they do. But where is the proof? Not vague claims of evil doings, not gut feeling, but proof?
“As our reader amd-linux (from Germany) put it earlier today:”
In response to a post by me, which like the previous one on this topic I referenced, you cannot bring yourself to reply to. Does logic make you nervous? Or is it just me? Oh well, at least you’re not pasting silly red text to my posts anymore. Somebody got called on that, eh?
Yes, because at the beginning of the post you say:
Right.
Chips B Malroy said,
May 14, 2009 at 5:11 pm
http://www.league2000.co.uk/e107_plugins/log/stats.php?4
Notice this is a UK site where we would expect Linux usage to be low. Seems to be a site about cycling mostly, which might skew the figures some, which way not sure. The approximate sum of linux users for this site (5200 hits or more) is approximately 6%. This is a lot more than the 1% from Net Apps.
Now Net Apps may not be the M$ Shill that I believe they are, but still, the evidence in their favor is not good.
Now Silverlight and Mono, and why M$ wants it. It more than just the patent club and trap I think, although that maybe a reason or the main reason. M$ wants us to think that GNU/Linux desktop use is tiny, tiny enough that home users will not try it. A percentage of home desktop market share is worth Billions to M$. So using and controlling a survey, like they have with just the facts sites in the past, is nothing new to this company.
While M$ wants us to think that Linux desktop use is still very tiny, ask yourself “why?” Wouldn’t Linux get more commercial apps and games(and drivers) for it if companies seen that it was much larger than Net Apps claims?
Even Steve Ballmer has said that Linux is the main competition of M$. Not Mac as much. So why Mono then, beyond the patent trap? Could it be, that M$ knows the real numbers of Linux users, and needs it silverlight running on the Linux desktops, in order to have any chance at all of competing with Adobe Flash? But yet at the same time, M$ doesn’t want to be seen as “porting” silverlight to Linux itself, so it uses Novell to do its dirty work to infect Linux with its crappy buggie software. Only place that silverlight is being used is Netflix in volume. Could it be because the CEO of Netflix is also on the board of directors of MSFT? Follow the money.
Gentoo User Reply:
May 14th, 2009 at 6:17 pm
Sorry, but assuming you had managed to stay on topic and not babble on about the Evil Mono and Silverlight and such, I just can’t bring myself to take anything written this way with any seriousness. Try again, it’s actually easier to just type “MS”, and people won’t think you’re a petty child with emotional problems who can’t express himself fully without that clever “i’m on the internets sticking it to teh man” spelling. Don’t worry, no one is going to confuse the acronym and think you’re referring to Multiple Sclerosis or Mayan Spaghetti. Not here at least. There’s enough context in what you’re saying that there’s no danger of that whatsoever
Chips B Malroy said,
May 14, 2009 at 7:11 pm
Its always a thrill to know that one has posted a comment that has perturbed the M$ Shills. In that way, we know when we struck a nerve with their shills, we are on the right track.
Roy Schestowitz Reply:
May 14th, 2009 at 7:18 pm
Yes, I’d just ignore him. The comment arrived from a witch hunter that does not even use GNU/Linux.
The Mad Hatter said,
May 14, 2009 at 8:53 pm
The biggest evidence is Microsoft itself. Why would Microsoft make a deal with Novell, if Linux wasn’t a significant player? Why would Novell pay Miguel and friends to work on Mono and Moonlight, if there weren’t enough users to make it a paying proposition?
Easy, they wouldn’t. Thus Linux market share must be higher than Nett Applications reports, or it wouldn’t be worthwhile.