> I'm not from Microsoft and it's been a few years, but I remember
> reading somewhere that the POSIX support Tekherov refers to ...
that almost sounds/looks like TECH.-HERO -- "kinda funny", thanks. ;-)
regards,
alexander.
yyyyyyyyyy@xxxxxxx on 05/16/2002 09:13:47 PM
Please respond to yyyyyyyyyy@xxxxxxx
To: yyyyyyyyyyyyyy@xxxxxxxxxxxxx
cc: yyyyyyyy@xxxxxxxxxxxxx, Alexander yyyyyyyyyyyyyyyyyyyy@xxxxx,
yyyy@xxxxxxxxxxxxxxxxxxx
Subject: Re: POSIX.C++
I'm not from Microsoft and it's been a few years, but I remember reading
somewhere that the POSIX support Tekherov refers to was only partially
compliant, which I wasn't happy about, when it was introduced in Windows NT
3 and was reported as a bit buggy. Either in NT 3.5 or 4, the code
optimizations for the NT kernel to make the Win32 subsystem more efficient
further broke it, I read, and it was decided to remove the POSIX subsystem
layer completely rather than fix it. Given at the time there was more than
one standard being called POSIX, as ISOs and IEEEs efforts have been
somewhat disjoint in their emphasis, I can't blame MS but I wasn't happy
about it, which is why I remember; I was hoping they'd make it as compliant
as plausible. At that time, I believe, is when those links would have
stopped working, as they were no longer relevant to the base NT products.
Mr. Zio, if he's so incline! d, can clarify if I'm inaccurate here and
would have to provide the timeframe when it was decided to add back a UNIX
subsystem as a separate extra cost product. I further suggest this
discussion be continued on the off-topic redirector, not here.
Cheers,
Mark
In a message dated 2002-05-16 10:34:20 AM Pacific Daylight Time,
yyyyyyyy@xxxxxxxxxxxxx writes:
Try reading http://www.microsoft.com/windows/sfu
-----Original Message-----
From: Alexander Terekhov [mailto:yyyyyyyy@xxxxxxxxxx]
----- 8< -----
BTW, [and sorry for somewhat off-topic stuff] in regards
to Win32-"(which lack true POSIX compliance)"...
Does anyone from Microsoft here know what happened with
that nice article about WinNT's *POSIX subsystem* and
POSIX standardization process (bits like "POSIX" != "UNIX", etc.)... after
I've posted the link pointing to it (with a few annotations) to
comp.std.c++:
|