From: utzoo!decvax!ucbvax!info-cpm
Newsgroups: fa.info-cpm
Title: Re:  source code?
Article-I.D.: ucbvax.453
Posted: Tue Dec 21 23:38:46 1982
Received: Wed Dec 22 06:40:46 1982

>From goldfarb.ucf-cs@Udel-Relay  Tue Dec 21 23:37:34 1982
To: Herb Lin <LIN@Mit-Mc>
Cc: info-cpm@BRL
Via:  UCF-CS; 21 Dec 82 3:58-EST
Via:  Udel-Relay; 21 Dec 82 7:12-EST
Via:  Brl; 21 Dec 82 7:25-EST
Via:  Brl-Bmd; 21 Dec 82 7:58-EST

CP/M's standard distribution from DRI includes a BIOS source for MDS-80.
DRI does not distribute source for the "proprietary modules" (i.e., CCP
and BDOS) since these are the heart of the system.  But the BIOS is a
different matter because it is the interface to the hardware of the
target system.  If all we had to run CP/M on were out of the box TRS-80's
or the like, then we wouldn't even need BIOS source.

Even the systems of a particular hardware merchant come in different
flavors, so rather than try to cover all possible contingencies in a
single BIOS (which is what Heath has tried to do), the BIOS distributed
with such systems usually is a "greatest common subset" variety;  the
source is necessary if the user is to do any customization or  add
any new hardware.  Some CP/M's like the old Lifeboat-Northstar CP/M
are distributed BIOS source for only the portion not related to disk
I/O -- I guess they think there is something about disk drivers
that is magic and needs to be protected.  

Think of what one would have to go through if each time he added a 
new component to his system, say a new serial interface, he had to go
to the place where he bought his copy of CP/M and ask the proprietor
for an update with the new hardware routines!
					Ben