From:	andy AT pswtech DOT com ("Mr. X")
Subject:	Unix, Win32 & Posix subsystems [was: TeX implementations?]
29 Oct 1996 22:07:17 -0800 :
Sender:	daemon AT cygnus DOT com
Approved:	cygnus DOT gnu-win32 AT cygnus DOT com
Distribution:	cygnus
Message-ID:	<3276DC93.4CA6.cygnus.gnu-win32@pswtech.com>
References:	<199610290633 DOT PAA15705 AT bird DOT fu DOT is DOT saga-u DOT ac DOT jp> <9610291053 DOT ZM27457 AT morgan DOT com>
Mime-Version:	1.0
X-Mailer:	Mozilla 3.0Gold (WinNT; I)
Original-To:	Graham <ggill AT morgan DOT com>
Original-CC:	gnu-win32 AT cygnus DOT com
Original-Sender:	owner-gnu-win32 AT cygnus DOT com

> between making NT look like Unix by talking to the win32 API and 
making NT look like Unix by talking to the POSIX API.

On the surface, it sounds like Graham is asking a naieve question,
but at another level he's not.  He assumes that the NT executive can
be thought of as policy-neutral, that the "protected subsystems" are
peers to one another in some sense.  They might have been -- and might
yet be -- that would be cool, but they are not.  

My observation is that the Posix subsystem was simply a half-hearted
attempt; a show necessary in order to win certain contracts.  The OS/2
subsystem, never more than a contingency plan.

IMHO, Microsoft designers never really grasped the concepts of
network transparency and policy-neutrality.  Others have said they
understood them and rejected them on marketing grounds.  The mish-mosh
of display primitives and fundamentals such as container objects in
the MFC is merely an example.  The internals appear to be in very much
worse shape.

But anyone who agrees with Andrew Schulman about undocumented
interfaces between Microsoft operating systems and their applications 
will assume that there are undocumented interfaces between the NT
executive and the Win32 protected subsystem.  

What's going on with that?
-
For help on using this list, send a message to
"gnu-win32-request AT cygnus DOT com" with one line of text: "help".

- Raw text -