On Feb 6, 6:48 am, George Kettleborough
<g.kettleboro...@xxxxxxxxxxxxxx> wrote:
> On 06/02/09 11:09, Roy Schestowitz wrote:
> > Why is Net Applications (Hitslink) changing its browser stats after publishing
> > them?
>
> >http://my.opera.com/haavard/blog/2008/09/04/why-is-net-applications-h...
> > Hmmmmm...
> > Let's see...
> > One of the executives used to work for Microsoft.
> > Net Applications uses a Microsoft stack for hosting.
> > Net Applications does not count traffic.
> > Net Applications admits its stats are flawed.
> > Net Applications keeps its methods secret.
> This is also evident in their OS stats page which reports "Linux" at
> just 0.8%. This is below estimates that were made some years ago.
> Compare it to W3Counter:http://www.w3counter.com/globalstats.phpand
> W3Schools:http://www.w3schools.com/browsers/browsers_os.asp
> George Kettleborough
The sites you listed above are the better web sites for checking
relative growth, but there are still flaws, not the least of which is
that they can't see behind the firewalls of corporations, so they
undercount those systems completly. For example, ALL 450,000 of IBM's
users appear as 1 user.
On the other hand, when I'm using a verizon cellular modem which hangs
up frequently when I'm idle, I might get 30 different IP addresses in
a single day.
When W3Schools had Linux at 2% Google's last published Browser survey
showed Linux at 7%
That was in 2002. The W3Schools site shows Linux growing from 1.5% to
almost 4% in that same period of time.
Note that in most of these Web Browser surveys the browser type is
X11 not Linux, while Windows and Mac have the brand as the first
entry. In several surveys you have to figure out the "other" catagory
and add that to the Linux total.
|
|
