Linked by makfu on Tue 23rd Oct 2012 17:19 UTC
Windows Peter Bright has written a fascinating and well researched look into the real architecture and general internals of WinRT that largely contradicts some commonly accepted marketecture myths propagated by Microsoft (namely that WinRT is a new peer, NT native subsystem and framework alternative to traditional Win32 that doesn't depend on traditional Windows frameworks). Given all the recent news and discussions regarding the future of Windows and the openness of the platform, understanding WinRT's actual underpinnings and technical place in the Windows stack will prove valuable.
Permalink for comment 539997
To read all comments associated with this story, please click here.
RE: Comment by Nelson
by dsmogor on Thu 25th Oct 2012 15:00 UTC in reply to "Comment by Nelson"
dsmogor
Member since:
2005-09-01

For me the main message with WinRT that OOP api is now the lowest level general purpose OS api in Windows. The experiences for making .NET libraries have been taken lower level, with COM simply selected as object ABI.

The most important question is whether MS will keep Metro apps away from plugging into old Win32 underpinnings. Some policy could at least be put in place to discourage resource strapped programmers from doing short circuits from old code bases. Can e.g win32 be used on Arm? Without that, eventually retiring Win32 won't be possible.

Reply Parent Score: 2