Path: utzoo!attcan!utgpu!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!mailrus!purdue!decwrl!ucbvax!UNC.BITNET!UNCSPL
From: UNCSPL@UNC.BITNET ("Scott P Leslie")
Newsgroups: comp.sys.atari.st
Subject: Re:      Info-Atari16 Digest V88 #484
Message-ID: <8811130450.AA28924@ucbvax.Berkeley.EDU>
Date: 13 Nov 88 03:20:00 GMT
Sender: daemon@ucbvax.BERKELEY.EDU
Organization: The Internet
Lines: 32

Hi,
     This is more about the TOS rewrite.
.
Why isn't MINIX a good idea?  Because it costs money from what I hear
and the average ST user isn't going to want to pay $100 to get the
new TOS.
.
Here are some features that I would like to see:
.
     - better support for hard drives
     - networkable device drivers
     - generalized math coprocessor support (not just the 68881)
     - alternate screen support (such as the new 19" monochrome)
     - mutitasking (true or juggling won't matter to me)
     - much better memory management
     - documented ways of doing certain things
          (like changing the mouse pointer redraw routines, etc...)
.
Well, I could go on and I have a list somewhere of other things that
I have thought up over the years, but I won't bore you.
.
One thing that deserves attention now is the generalized math
coprocessor support.  There should be a standard vector for
interrogating mc availability and access methods.  If there was
then people without a mc would have a set of floating point routines
written in 68000 accessed by this vector.  However, the people with
the math coprocessor would still get almost full speeed out of the
arrangement.
.
Scott P. Leslie (UNCSPL@UNC)
.       Jax
(NEWD Software : Nothing Else Will DO)