Android devices outselling Windows devices by nearly four to one. But Windows 10 is clearly on the rise and more of us will soon be in the OS-as-a-service world, just where Redmond wants us to be.
This is a glaring example of a logical fallacy along the lines of: frogs are green; I am green therefore I am a frog.
An increase in relative terms (market share) does not mean an increase in absolute terms. Indeed, as the market share of mobile devices increases, the number of people using desktop machines decreases (absolutely and relatively). So as people switch from Windows to IOS or Android, and they are, the market share for Windows 10 might go up even if fewer people are using it. That said, 1 year in and even including the compulsory distribution on new machines (with the odd exception) and strong arm update tactics, the uptake of Windows 10 is far from impressive.
From my own numbers I'm watching a growth in mobile from around 15% in 2015 to over 20% in 2016. But don't believe me: have a look at Akamai's data, which El Reg persistently fails to refer to, along with its own data.
I have no idea why Mr Sharwood thinks data from US government websites is any more representative of global trends than data from El Reg. I suspect the answer is that he simply doesn't have access to El Reg's stats. This would be very poor for a tech website if it were true.