Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!tut.cis.ohio-state.edu!pt.cs.cmu.edu!andrew.cmu.edu!eb1z+
From: eb1z+@andrew.cmu.edu (Edward Joseph Bennett)
Newsgroups: comp.sys.mac.programmer
Subject: Re: Why not Mach as version 8.0
Message-ID: <kZyS8a200WBLQ1mmRV@andrew.cmu.edu>
Date: 11 Mar 90 05:29:10 GMT
References: <39127@apple.Apple.COM>,
	<1236@swbatl.sbc.com>
Organization: Class of '91, Carnegie Mellon, Pittsburgh, PA
Lines: 30
In-Reply-To: <1236@swbatl.sbc.com>

>Okay, perhaps someone can answer me this question:  why not use an
>implementation of Mach as verrsion 8.0 of the MacOS?  It would seem that you
>could preserve the shared memory areas via Mach's shared virtual memory
through
>VM inheritance (e.g. don't copy-on-write but inherit a shared chunk of VM).
>This would allow the large majority of Mac programs which do twiddly things
>in the system heap and such to still run correctly.

I don't see why not?
In the February issue of Cursor, CMU's acedemic computing magazine,
there is an article on MacMach " We are currently developing software
that will allow MacMach users to run Macintosh applications under Mach.
This system will allow the Macintosh Operating Sysytem and Toolkit to
startup in a single MacMach process, and run either the Finder or
Multifinder. To the user, a Macintosh running MacMach will appear no
different than a Macintosh running  the Macintosh Operating System,
except that the MacMach user will also have acess to UNIX and Andrew
applications. This system is in the experimental stage, but may be
available in a limited form some time this spring. "

Now since Apple is funding this project..., well you can infer the rest. 

Such a system that could exploit the power of Mach and UNIX but still
run all Mac applications with the Mac interface would make the high end
Macs formidable contenders in the Workstation Market. 

Funny, As a CMU student I may be able to test such a system BEFORE I get
my hands on system 7.0.

Ed