python 2.7.12 on Linux behaving differently than on Windows

[eryk sun]

On Wed, Dec 7, 2016 at 9:43 AM, Chris Angelico <rosuav at gmail.com> wrote:
> Exactly. Back in the 1990s, we had the beginnings of Windows NT, which
> was designed for servers. It had (if I recall correctly) no concept of
> file/directory permissions, little or no notion of process privilege,
> and definitely no way to restrict one process to X amount of memory
> and Y amount of processing.

Do you have a name or a link for this system that pre-dates NT? At the
time I thought Microsoft only had MS-DOS and OS/2 (and maybe still
Xenix?). I know they hired Dave Cutler (the designer of VMS) and
several other DEC programmers in late 1988 to develop NT OS/2. It took
them 5 years, and along the way the design switched from NT OS/2 to
Windows NT (DOS-based Windows had surged in popularity with the
release of Windows 3.x, and the business relationship with IBM fell
apart). They relegated the OS/2 and POSIX subsystems to the console
only, without GUI support (similar to the new Linux subsystem).

> Since then, Windows has progressively
> added features that Unix has had for decades, because those features
> are critical. Those who don't understand Unix are forced to reinvent
> it, piece by piece, badly, and then finally rip it out and start
> over...

NT was designed to run a Unix subsystem and meet U.S. DoD security
requirements, so it's naturally going to share a lot of capabilities
with Unix. But the similarities are superficial; the design decisions
are fundamentally different in many ways. (The differences are more
obvious in native NT and kernel development.) No one could say the NT
design team was copying Unix. OTOH, how much they pilfered from their
previous work on VMS is debatable. IIRC, Microsoft ultimately reached
a settlement out of court with DEC on this issue, which was the main
reason NT was ported to the DEC Alpha processor in the 90s.