Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!apple!apple.com!dbrown
From: dbrown@apple.com (David Brown)
Newsgroups: comp.os.minix
Subject: Re: MacMINIX
Message-ID: <8582@goofy.Apple.COM>
Date: 6 Jun 90 21:34:19 GMT
Sender: usenet@Apple.COM
Organization: Orion Network Systems
Lines: 29
References:<1458@lectroid.sw.stratus.com> <6785@star.cs.vu.nl>

In article <6785@star.cs.vu.nl> ast@cs.vu.nl (Andy Tanenbaum) writes:
> Apple's attitude is that you are a user and you can be trusted to
> point with the mouse.  How the disk controller, clock, video,
> keyboard and DMA chips work is none of your bloody business.
> Kind of hard to write a native OS for a secret machine.

But it also allows Apple to make significant changes to the underlying 
hardware (witness the recent IIfx) without breaking existing software.  
Each new machine and system software seems to break a few things, but 
think about how much worse it could be if Apple documented the hardware 
and developers wrote to those specs.  And you could end up with different 
versions for each different CPU.

(This is my own opinion; I don't know what Apple's official justification 
for secrecy is.  In general I dislike secrecy; I would love to see 
everything documented, but it would have to be on on a per-CPU basis, 
which could eat up a lot of resources.  I'd rather those resources go in 
to developing and testing new machines.  Apple seems to have its hands 
full just keeping Inside Mac, which documents the ROM calls, up to date; I 
know that many Mac developers would love to see a revised single version 
of Inside Mac (rather than having to refer to Volumes IV & V for recent 
info).


David Brown        415-649-4000
Orion Network Systems
(a subsidiary of Apple Computer)
1995 University Ave. Suite 350
Berkeley, CA 94704