Path: utzoo!utgpu!watserv1!watmath!att!tut.cis.ohio-state.edu!snorkelwacker!usc!zaphod.mps.ohio-state.edu!sdd.hp.com!decwrl!shelby!neon!minnie.Stanford.EDU!philip
From: philip@minnie.Stanford.EDU (Philip Machanick)
Newsgroups: comp.sys.mac.misc
Subject: Re: Where does UNIX fit in a graphically-based computer world?
Message-ID: <1990Sep10.193027.14774@Neon.Stanford.EDU>
Date: 10 Sep 90 19:30:27 GMT
References: <1990Sep10.063504.27780@world.std.com> <1990Sep5.202652.700@sun.soe.clarkson.edu> <1990Sep5.224940.19185@world.std.com> <1990Sep6.161554.28923@ux1.cso.uiuc.edu>
Sender: news@Neon.Stanford.EDU (USENET News System)
Reply-To: philip@pescadero.stanford.edu
Organization: Computer Science Department, Stanford University
Lines: 58

In article <1990Sep10.063504.27780@world.std.com>, boris@world.std.com
(Boris Levitin) writes:
> dorner@pequod.cso.uiuc.edu (Steve Dorner) writes:
> 
> * DEC: I've nothing against it; it keeps claiming a price/performance
advantage
>        over HP and Sun RISC machines, but for some reason its penetration 
>        currently seems to be low. As far as I understand, the
VAXstations use a
>        proprietary RISC chip that requires recompilation of software
running on
>        other UNIX boxes, and right now there are still few GUI-based programs
>        running on the former compared to the latter.
Actually, that's DECstation. The Vaxstation uses the Vax cpu. The DECstation
uses the MIPS processor. While it hasn't sold in the same volume as Sun's
SPARC, the MIPS is used in a number of other workstations, including MIPS's
own, and Silicon graphics.
> 
> * NeXT: I've just come back from MacWorld Expo/Boston, ...
>         ...(it's slow, has 2-bit video and costs the Earth).
>         Just like the first Mac, it's a lacking implementation of a
good idea.
Rather more like the Lisa, I would say.
> 
> * IBM: To quote P. W. Botha, it must adapt or die.  I trust it will adapt,
>        but whether with very interesting products, I don't know.
Gosh. I didn't know anyone still remembered PW Botha. Not even South
Africans...
> >It is also relevant to mention that the GUI vs. UNIX thing cuts both ways;
> >not only is UNIX getting a GUI, but GUI-heavy machines are finding they
> >need operating systems.  Witness OS/2, Macintosh System 7, and the new
> >Windows.  It is at least as valid to ask whether these operating systems
> >will be any good as it is to ask if the GUI's on top of UNIX are any good.
> 
> >My own feeling is that it is much easier to put a good GUI on UNIX than it
> >is to slip a REASONABLE operating system under an existing application base
> >(the game currently being played in the IBM PC and Mac worlds).
> 
> Amen.  I'm running into the toy aspect of the current version of the Mac OS
> right now.  I hope Apple gets serious about giving the Mac a real operating
> system, based on UNIX, free...
Unix needs a vast amount of cleaning up before it can be dumped on an
unsophisticated mass market. I agree with the principle that Apple could do
what they've done with AUX: support a Mac compatibility box in a "proper" OS.
However, as we've seen with OS/2 and System 7, designing a fresh OS is a
non-trivial task (even with 2 opposite ends of the scale on
"compatability box" -
in OS/2, it's not intended to be the main mode of operation; on System 7, it's
the _only_ mode of operation). I think Apple can learn a lot about OS
design/implementation from AUX, but I'd prefer it if the lessons learned went
to creating a newer better OS (but still with _both_ Mac and unix compatability
capability).

Maybe after they finish System 7, they will have some idea of how big a job
this will be...

Philip Machanick
philip@pescadero.stanford.edu