Re: OFFTOPIC: Regarding NT vs Linux

Chris Wedgwood (chris@cyphercom.com)
Sun, 21 Sep 1997 14:54:51 -0400


From: Gerard Roudier <groudier@club-internet.fr>
To: Alan Cox <alan@lxorguk.ukuu.org.uk>
Subject: Re: OFFTOPIC: Regarding NT vs Linux

[...]

(1) STUPID people who are happy to lose about 50 % of the performances
due to the bad implementation of the O/S and its drivers.
(2) People who have PARKINSON's desease and think that using a mouse
for interfacing an O/S can cure for.
(3) People who want to always apply PARKINSON's law.
(4) Bill lovers.

Crap.

I hate M$ more than anyone. I actually used to just dislike them, but now I
hate them. I have even turned down work based upon this M$ loathing. (That
said, I own several copies of NT, MSVC++, BC++, Office, etc).

But...

What does M$ offer that no unix does at the moment? (MacOS offers most of
this too...)

+ A _reasonable_ gui that 99% of people use, with controls that 99% of
people have an can use.

It standardized. I can write an MS app and assume they will not have to
install lots of widget/control libraries, etc.

Even key-bindings are fairly standardized these days. (Most windows
programs know Control-X is cut...)

There are other standardized APIs that people by and large use. Things
like TAPI, DirectX, etc. Most of there are fairly well documented which
helps both vendors and software developers.

+ NT's performance isn't as bad as some people would like to believe. For
some things its actually quite good. Certainly not at 50% of Linux anyhow.

There are also some kewl things like TransmitFile where you can send a
data stream out of a socket without having to copy the data in chunks to
and from user-space.

Sure, Linux could have this.... but it would it be useful if no other unix
had this.

As is then current NT web servers actually perform very well.

There is something to be said for the efficiencies of dictatorial
decisions. Linus "I don't like select updating timeout, even if it is the
Right Thing to Do and POSIX doesn't say we shouldn't do this" Torvalds
might even agree here :)


[...]

Bill's O/Ses have success because software/hardware companies trust M$$$
perennity. The main enhancements of NT have been to add to Bill's
Windows System features that existed since more than 15 years on Unix
and proprietary systems.

There is also a degree of certainty about that will be available to end
users a year down the track. Still, things like WinG do happen.

If a vendor writes a device driver for Windows95 - generally it will work
with a newer kernel. The same cannot be said for linux. Things like the
module API and stuff are too volatile/sensitive to changes.

Linux needs a clearly defined API for many things. Yes, this lock you into
support old backwards crap. SO, you do so. Then a year or so later, you
release a new version a throw away backwards support to an extent (Al la
SunOS -> Solaris).

I'm not sure that you can solidify all of the APIs under unix either. Mainly
because everyone has their own equally legitimate ideas of how things should
work, and how they should look. But I can hope, can't I?

Users have trusted CRAPDOS and CRAPWIN for 15 years waiting for Bill to
decide to offer them 20 years old features. We have all been done, and
it seems that this swindling is not about to stop.

[...]

Perhaps most people are stupid and they should be killed at birth. How do we
tell a 'Linus Torvalds' at birth from a 'Bill Gates'? (Or a Bill Clinton?)

Just my opinions, 'tis all.

-Chris