Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!watmath!clyde!rutgers!ucla-cs!zen!cory.Berkeley.EDU!iverson
From: iverson@cory.Berkeley.EDU.UUCP
Newsgroups: comp.sys.ibm.pc
Subject: Re: UNIX vs. OS/2
Message-ID: <3835@zen.berkeley.edu>
Date: Sat, 19-Sep-87 05:56:01 EDT
Article-I.D.: zen.3835
Posted: Sat Sep 19 05:56:01 1987
Date-Received: Sun, 20-Sep-87 14:32:28 EDT
References: <494@parcvax.Xerox.COM> <961@looking.UUCP> <498@parcvax.Xerox.COM> <3834@zen.berkeley.edu>
Sender: news@zen.berkeley.edu
Reply-To: iverson@cory.Berkeley.EDU.UUCP (Tim Iverson)
Organization: University of California, Berkeley
Lines: 59

In article <3834@zen.berkeley.edu> c60b-ia@buddy.Berkeley.EDU.UUCP (Sugih Jamin) writes:
[in response to an article comparing UNIX vs. OS/2 to MSDOS vs. CP/M]
>The difference is:  MS-DOS can do *much* more than can CP/M.

Not true.  When it was first released, MSDOS couldn't fight its way out of
a wet paper bag (it still can't, but then it doesn't have to).

>  While OS/2
>won't be that much more powerful than UNIX, if at all.

You seem to have it wrong here (as well as not having listened to the
recent articles).  UNIX, now, is much a more capable and complex OS than
OS/2 will be when it is released.  It has taken UNIX more than 10 years
to get to where it is today.  It will undoubtably take OS/2 10 years to
reach the point at which UNIX is today.

Unfortunately, OS/2, when it finally reaches the distributor's shelves,
will already have greater popularity among PC users than all of the
flavors of UNIX combined.  The big corporations will buy it because it
has been baptized and blessed by big blue (how's that for alliteration!)
and the little guys will eat it up because it allows them to keep abreast
of current innovation while maintaining their current investment in software,
which is considerable.  This is the marketing point of view, and the one
that will win out in the wide world of users.

Developers are a whole different story; they don't need compatibility
since they generally posses the expertise necessary to use a system that
will provide maximum programmer throughput.  Since OS/2 allows real-mode
programs, it will be inherently unsuitable for this task - no protection
means no post-mortem debugging.  Industrial strength development demands
an OS that is robust enough to maintain its poise in the face of a stranger
set of bugs and wild pointers than MicroSoft can dream will ever happen
(their concept of well-behaved programs is a joke, as if programs were
puppies to be trained to use the paper!).

For development, then, UNIX seems to be the OS of choice, mostly because
of its current state of availability.  Especially since 386-UNIX will be
able to take advantage of the 386's virtual-86 capability to provide
a protected version of OS/2's real-mode, allowing developers target their
products for all levels of the PC-clone market - 8088 through 80386.
OS/2 won't even run on the 386 (in the sense that it won't support any
programs that use all 32bits of a register - might as well buy a 286).

Then there's the philosophic side to the matter, which all of ya out
there in netland have been arguing - you know, the one about whether
the OS should hold the user's hand and whether it should hold the right
hand or the left?  Well, there were some awfully good arguments, and
I agree with all of them - if the computer can figure it out for itself,
then it should, and let the user get his work done.  But, (y'all knew
there'd be a but, didn't you?) that wasn't the point of the original
article and that was that this whole big OS thing was decided on sales
and marketing issues, not philosophy.  And it has already been decided,
while the masses gobble up OS/2, we can dine on UNIX, or in the immortal
words of Marie Antoinette, "Let them eat cake."


- Tim Iverson
  iverson@cory.Berkeley.EDU
  ucbvax!cory!iverson