Path: utzoo!attcan!uunet!pyrdc!pyrnj!rutgers!ucla-cs!acm
From: acm@valhalla.cs.ucla.edu (Association for Computing Machinery)
Newsgroups: comp.sys.atari.st
Subject: Re: A proposal--TOS Replacement Project
Keywords: TOS, replacement project, STOS
Message-ID: <17702@shemp.CS.UCLA.EDU>
Date: 9 Nov 88 05:07:13 GMT
References: <700@sdcc15.ucsd.edu>
Sender: news@CS.UCLA.EDU
Reply-To: acm@cs.ucla.edu
Organization: UCLA Computer Science
Lines: 61

In article <700@sdcc15.ucsd.edu> pa1132@sdcc15.UUCP () writes:
[...]
>brain-damaged TOS.  The new operating system should be called
>"STOS", short for "SmarTer Operating System", meaning an OS smarter
>then TOS.  Any comments?

I have a comment.  Why not adopt MINIX as that new OS?  

Some people could possibly write a "gem" program for it that would
allow regular ST software to run compatibly.  If this could be done
then I think MINIX should be first on the list since:

	* MINIX can be succesfully used with other processors.  I remember
	  a case a while ago in which someone using MINIX replaced the 68K 
	  with a 68020.  There wasn't much of a speed increase, but then I 
	  don't think he went "all the way" with the hardware (i.e. a faster
	  clock for the CPU and for the separate board of memory and/or an 
	  SRAM cache).  I guess if you had to run true-blue GEM programs you 
	  could just toggle a switch.

	* MINIX multitasks.  I have thought about it for a while, and
	  have come to the conclusion that in order to implement powerful
	  networking for the ST, a multitasking OS would make things 
	  much, much easier.  Of course there are other benefits to 
	  multitasking, but that is an exercise left to the reader.

	* MINIX already includes the source!  That means that if there is 
	  a bug, YOU have the chance to fix it without waiting on anyone.
	  Having the source to the OS also opens up other worlds of 
	  possibilities, like implementation of sockets (am I dreaming?),
	  an added dimension to debugging (e.g. if you want to find out what
	  the OS does during disk writes, you just look and read the 
	  comments!), a boost towards being able to make it TOS compatible,
	  and many other advantages.
	  
	* MINIX is robust.  This is a second source opinion as I don't have
	  MINIX yet.  However, most of the bug reports I've seen in
	  comp.os.minix have been pretty piddly-type things.  Shake well
	  and give time for setting of contents.

	* etc...

For your convenience, I include here what someone in comp.os.minix had to say:
 
>     I have been using/playing with Minix for the ST for a couple of weeks
>  now, I think that it beats gemdos by a mile, even though it, at this point,
>  does not support all of the st's hardware.  I have a few questions I hope
[rest deleted]

So far I have not come across someone that has HATED it.  Will anyone who has
MINIX already care to comment or provide the disadvantages?


Plinio Barbeito
P.S.
It really would have been best for me to post something like this AFTER I had 
my copy of MINIX, but as long as you were asking for people's 2 cents...
---
UUCP:  ...!{...}!ucla-cs!acm
ARPA:  acm@cs.ucla.edu
VOICE: (213) 825-5879, 825-7597