Micro$oft Biz'droid Lusers (was: OT email response format)

From: Richard Erlacher <edick_at_idcomm.com>
Date: Tue Apr 23 10:57:51 2002

Your comment that one should use Windows for office work is well taken. It
does that quite well, and other OS' feeble attempts at it realy don't compare.
That's probably the primary reason for Windows' success.

The fact that *NIX has focused on other things is why it's not popular with
folks engaged in other pursuits.

However, because of Windows' amazing popularity (despite its sometimes glaring
weaknesses) other application types are finding their way into the Windows
environment. Because of the ghastly results produced under UNIX for the EDA
community, not to mention the 2-orders-of magnitude-higher prices, this is
welcomed by those of us who rely on EDA tools to make our living. The tools
we had under UNIX used to break down just as often as under Windows, but they
cost WAY more, and one couldn't blame the OS.

All the *NIX sophistry is what has cause the decline of *NIX popularity in
general, and the MS-hating doesn't do a thing to help the *NIX community.

All the MS-haters' rants do is shore up the belief that it's just sour grapes.
If one could take one of today's high school grad's and plunk him/her down in
front of a *NIX box and be able to get a little useful work from him/her THAT
SAME DAY, which one certainly can do with Windows, then *NIX would be much
more widely used.

Dick

----- Original Message -----
From: "Bob Bramwell" <bob_at_copenhagen.cuug.ab.ca>
To: <classiccmp_at_classiccmp.org>
Sent: Tuesday, April 23, 2002 12:22 AM
Subject: Re: Micro$oft Biz'droid Lusers (was: OT email response format)


> I didn't catch the posting that seems to have started all this, but
ignorance
> never seems to stop anyone *else* from stating an opinion, so here's mine
:-)
>
> A few years back I spent a little time working for a small company whose
primary
> focus was Unix system software. They did some Windows work, and this was
their
> eventual downfall because they were bought out by McAfee who didn't have a
clue
> what to do with the Unix stuff and weren't interested in finding out.
>
> I was the last remaining employee of the original company. For most of the
last
> year of its operation I did pretty much all the system administration,
> lightweight hardware maintenance, bug fixing, and many of the enhancements
to
> the two Unix products that had been the mainstay of the company. I forget
> exactly how many machines we had and exactly what they ran, but it was
something
> like:
> Hardware OS Notes
> ======== == =====
> HP/Apollo 700 HP/UX 9.x
> HP/Apollo 700 HP/UX 10.01 + 10.10? dual boot
> DEC Alpha 2000 (?) OSF 1
> Motorola Mxxx Motorola Unix v.X forgotten machine
> Motorola Myyy Motorola Unix v.Y & OS version names
> SUN Sparc 5 Solaris 2.5.1
> SUN SLC Solaris 2.5.1
> SUN 3/60 SunOS 4.1.1
> IBM RS6000 AIX v.X & v.Y forgotten versions,
> dual boot
> Intel SCO Unix
> Intel Dynix
> Intel ?nix more memory loss
>
> I am missing a couple of machines and maybe one more OS. Before I started
with
> the company I had never even heard of some of these variants. The point of
all
> this is that the same code base built on ALL these things. Sure, we had a
> "portability" library of our own to make it easier, there was a fair amount
of
> #ifdef stuff, and we had to tinker with GNU autoconf to get the builds
smooth,
> but I could manage ALL of it. We used Emacs and gcc/gdb on all the
machines,
> and a freeware distributed backup package (Amanda) to run complete backups
of
> everything automatically (well, OK, I had to change the tapes manually). Of
> course, with the network applications that typically come with Unix (telnet,
> ftp, NFS, X Windows, etc.) I could sit at my own desk and get at every
machine
> conveniently and run a parallel build on all the machines. That would have
been
> completely impossible with the Microsoft (bundled) products available at
that
> time.
>
> Now, I've watched people trying to do software development on Windows
systems
> over the years, and I've done a little myself here and there. Every time
there
> is a new service pack or OS release there are wailings and rendings of
clothes
> on all sides. New licenses have to be bought for compilers, new versions of
the
> network backup client have to be purchased, very often a whole new machine
has
> to be bought, the sys admins have to do all sorts of black magic to get the
new
> machines to "play nice" with the old ones on the network, and a whole new
set of
> OS idiosyncracies has to be mastered. I have never observed anything
similar in
> Unix development shops (well, OK, the more organised ones, anyway). Oh, and
> even *less* upheaval takes place in VMS shops BTW, but VMS has other
problems.
>
> So:
> 1. it is perfectly possible to write highly portable software, and once the
> initial portability setup is done the rest is not particularly hard.
> 2. there is a great deal of very useful "free" software for Unix out there
in
> net land, much of which you would have to pay big $ for on Windows
> 3. there is NOT very much "business" software (word processors,
spreadsheets,
> contact management etc.) for Unix because the kind of people who write
free
> software aren't generally very interested in "business" stuff, and the
kind
> of people who write commercial software don't think there is much of a
Unix
> market for it (since they priced themselves out of it a while back).
> 4. if I were developing software for "back end" applications (servers,
> networking, etc.) I would certainly do it on Unix in preference to
Windows,
> and in this respect the market mostly agrees with me.
>
> By all means use Windows for your office work. That's about all it's good
for,
> and on the whole it does it fairly well.
>
>
> Bob Bramwell Snail: 60 Baker Cr. NW | If I die in war you remember me;
> ProntoLogical Calgary, AB | If I live in peace you don't.
> +1 403/861-8827 T2L 1R4, Canada | - Spike Milligan (1919 - 2002)
>
>
Received on Tue Apr 23 2002 - 10:57:51 BST