🚀 go-pugleaf

jjs@dostoevsky.ucr.edu (Joe Sloan) writes:
>In article <DCL9rv.8yA@world.std.com>, jim frost <jimf@world.std.com> wrote:

>>Whether or not NT has been selling well depends on what you think its
>>intended market to be.  Certainly it's fallen flat as a mass-market
>>desktop system, but in my opinion it was never intended to be such.
>>The feature set is all wrong.  If you consider it to be a UNIX
>>workstation competitor, as I do, it's been doing quite well: last year
>>it had more than double the sales rate of the most popular UNIX
>>variant!  Even if its sales rate remains flat from 1994 it will match
>>the installed base of the second largest UNIX variant this year.  Best
>>available information indicates that 1995 sales will be significantly
>>better than 1994, though, so most likely NT's installed base will
>>greatly exceed that of *any* individual UNIX variant by the end of
>>this year.

>Microsoft cheerleading aside, let's look at this logically:
[...]
>windows nt has it's place, and we have nt machines here; but to think
>that nt could even begin to replace the unix machines will result only
>in a knee-slapping, brings-tears-to-your-eyes fit of laughter...

I'm not a Microsoft cheerleader by any means, but I think you're
missing the point.

NT won't replace UNIX servers in the near future -- there's too much
they do that it can't, as you point out.  On the other hand it's worth
remembering that the lion's share of UNIX systems that ship today are
desktop systems, not servers.

How many of the services UNIX provides are *really* necessary for
desktop systems?  Of those services, how many are unavailable for NT,
and how many have workable (or even superior) counterparts?

Most of those services aren't necessary, and any that are either have
add-ons or complete replacements (and most often both) available.  If
you want NT to talk to your UNIX systems, it's easy to do.  I know,
because I've been doing it full-time for more than 8 months.  All it
took was Samba and an X server.

So it's possible.  Why would anyone do it?  Software.

NT allows you to tap the vast PC software market in a way that no UNIX
can.  Good, cheap software is *extremely* compelling!

Some high-end software won't run on your NT box, or won't run
adequately.  Such software typically doesn't run all that well on
desktop UNIX systems either; in those cases it becomes cost-effective
to buy a big system and share it.  That works as well with NT clients
as it does with UNIX.

People's attitudes towards NT penetration into the UNIX market are
highly reminiscent of the way VMS people thought about UNIX in the
mid-1980's.  UNIX couldn't possibly take over the VMS market, because
it didn't do all these things.  It turned out that UNIX did do most of
them, although not always in the same way.  Enterprising software
developers filled in the holes.  What triggered the change was quite
simply cost-effectiveness; you could put together a working UNIX
system for a lot less money than a VMS system (and it was faster to
boot).  Today VMS is all but a memory.

I'm a big UNIX fan and have been for more than a decade.  There are
probably no more than a handful of people reading this message who
have more UNIX experience than I.  Regardless of how much I like it, I
recognize that it's not a terribly good desktop system -- it's too
complicated for what most people want to do, and software has remained
either very expensive or quite primitive.  It is the workstation OS of
choice not because it was what everyone wanted, but because there was
nothing else available that had anywhere near the same
price/performance.  Well, those days are over.

NT isn't going to kill UNIX, at least not in the near future.  We're
not going to wake up tomorrow and see an all-NT workstation world.  It
doesn't work that way, it happens one machine at a time, and it's
worth noticing that you already have a couple.

jim frost
jimf@world.std.com
-- 
http://world.std.com/~jimf