Newsgroups: comp.sys.mac.misc
Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!usenet
From: Garance_Drosehn@mts.rpi.edu
Subject: Re: Will the bugs in 6.0.7 be fixed (i.e. will there be a 6.0.8)?
Message-ID: <l5#gv0k@rpi.edu>
Nntp-Posting-Host: gilead.its.rpi.edu
Organization: Rensselaer Polytechnic Institute
References: <YEE.91Mar28122900@katana.osf.org> <1991Mar30.172229.193@uokmax.ecn.uoknor.edu> <7720@idunno.Princeton.EDU>
Date: 2 Apr 91 00:25:35 GMT
Lines: 27

In article <7720@idunno.Princeton.EDU> 
           bskendig@bonnet.Princeton.EDU (Brian Kendig) writes:

 ...(lots of reasonable stuff that I'm skipping over)...

>Apple once told developers that they could safely assume any 020- or
>030-based Macintosh would have a math coprocessor in it.  Apparently
>they warned people about their change in policy while they were
>designing the LC and IIsi, but by then people were already using
>applications which assumed the coprocessor.

I'm not too sympathetic with this particular example.  Whatever *Apple* said
about *Apple* products, there were third-party folks that were selling
accelerator boards for years before the LC and the IIsi showed up.  On those
accelerator boards you *can* have a 68020 with no floating point coprocessor. 
I bought one such board in 1989, and that was because the company was closing
out the line (ie, it wasn't some new and novel thing when I bought it). 
Certainly the larger companies must have had some customers who had such
machines, in which case they would have had plenty of time to write software
which wouldn't crash on a coprocessor-less 68020 machine.

It particularly bugs me when Microsoft tries resort to this excuse.  They don't
seem to mind ignoring Apple's guidelines whenever the mood strikes them.  But
on this one they want to pretend they were somehow forced into the problem
because of something Apple said. 

Garance_Drosehn@mts.rpi.edu
