Mary Jo reports that Windows 8 sales are roughly on par with Windows 7 sales. Which is good news for Windows 8, because Microsoft said (at the time) that Windows 7 was the fastest selling OS to that point.
She also points out that actual usage of Win8 isn’t terribly high at this point – which isn’t at all surprising (see my blog post on if Windows 8 is a success).
The real value of the numbers just provided by Microsoft is that they are an apples to apples comparison between Win7 and Win8, and that they demonstrate that Win8 is following roughly the same track as Win7 in terms of production and sales.
That’s good news, given that Win7 is (by nearly any measure) extremely successful, and is considered by many people to be the best OS Microsoft has released. Windows 8 on an x86 machine can basically be viewed as a faster version of Windows 7, plus the ability to run WinRT apps, and so I pretty much think of Windows 8 as a slight improvement over the already excellent Windows 7.
As Mary Jo notes, we don’t know if the 100 million figure includes Windows RT. At this point I’m not sure if that really matters – at least not from a business app dev perspective. Windows RT can only run WinRT (Windows Runtime) apps, and the WinRT dev platform is too new and immature to risk targeting it when building large enterprise apps (not to mention the side-loading cost issues).
At this point most organizations appear to be building new smart client apps using WPF, and of course they continue to maintain a great many Windows Forms apps. The strength of Windows 8, as I see it, is that it remains an extremely relevant and potent business app platform via its desktop mode, which runs Win32/.NET apps at least as well as its predecessor.
If Microsoft resolves the side-loading cost issues so licensing and deployment becomes reasonable for small, medium, and large organizations I do think WinRT has a reasonable shot at being the successor to Win32/.NET for business developers. In another version or two it should stabilize and mature to the point that it is pretty comparable to WPF, and thus is attractive and useful to C#/XAML developers. That’ll probably take a couple years, which is also the timeframe that corporate IT groups will probably be willing to consider upgrading from Windows 7 to Windows 8.
In summary: good Windows 8 sales today means that betting on WPF for smart client development should be pretty safe, and will hopefully have a decent migration path to WinRT in 2-3 years.