Re: Innovations

["Bruce R. McFarling" <ecbm@xxxxxxxxxxxxxxxxxxx>]

On Fri, 3 Oct 1997 11:14:04 -0400 (EDT), CHU@xxxxxxxx wrote:

>Mason Clark wrote:
>>   Oh, by the way, the first generally useful modern computer programming
>>   language, Basic (an academic invention), was introduced to industry by a
>>   couple of guys with a little company whose egg grew into a monster: MS.

>Just a few corrections.  The first high level languages to be popularly
>used were COBOL and FORTRAN, variations of which are still employed
>within academia and commercial data processing plants everywhere.
>The first academically influential modern languages were ALGOL and PASCAL.
>IBM introduced their own massive language: PL/1.

>All this is way before Microsoft and the microcomputer even existed.

	And Basic was introduced on minimcomputers before Microsoft existed
as well.  AFAIR, the original design intention of BASIC was to develop
a language for teaching programming that had simple syntax, so that
most of the course could concentrate on programming rather than syntax,
and was interpreted, to reduce the time lost from the 'code, submit,
compile, syntax error, code' (repeat as necessary) 'code, submit,
compile, runtime error, code' (repeat as necessary) cycle.
	Microsoft developed a Basic for early kit microcomputers, and
then Basics for CP/M machines and the early 6502 and Z80 home computers
(including the first microcomputer to sell in the millions).

>Furthermore, the BASIC that was incorporated into the initial IBM PC
>was a severely crippled version which made it very difficult to do useful
>work in. MS-DOS was descended from CP/M.  For many versions, the system calls
>from CP/M remained within MS-DOS.  In other words, there was little
>that was innovative in the software side of the IBM PC.

>What WAS innovative was the marketing, pricing, and the hardware.
>Suddenly people can run their own machines and program them instead of
>begging the data processing department to get something done.

	There were a few changes from CP/M, that some will say were due
to another operating system that Microsoft stole ideas from.  The
primary technical innovation, thought, was that MS-DOS was a more
accomodating platform for moving programs for CP/M for the 8080/Z80
than Digitial research's CP/M 86. *Combine that* with the licensing
arrangement that made MS-DOS the default choice that came with the
purchase of a disk drive and MS-DOS became the OS of choice for
business software transported from CP/M.  Along comes Lotus 1-2-3,
and the rest, as they say, is history.
	As a side-note, one of the reasons that CP/M-86 was a harder
target for existing CP/M-80 and CP/M-Plus programs was that, based
on experience with the multi-tasking MP/M-80, CP/M-86 was designed
to reduce the roadblocks to multi-tasking.  MS-DOS didn't bother
with that, and so it took almost a decade for Wintel machines to
get multi-tasking capabilities approaching capabilities available
in the mid-80's with machines like the Amiga.  How did Digital Research
get multi-tasking with slow, 8-bit processors?  Easy: add processor
cards to the CP/M system.  CP/M is a terminal oriented system, so
you just plug a terminal into its own processing card with its own
64K (or more) of RAM, and use the multi-processing system to control
the sharing of resources like disk drives and printers.

	As far as language, the big shift that occured around the
time of the IBM-PC was the development of a 'shrink-wrapped' software
market that made knowledge of programming languages an optional
rather than a required aspect of operating microcomputers.  And
in the shrink wrapped market, it was Borland's Turbo Pascal,
with its integrated editor / compiler development environment
that established the norm that later language venders had to
follow.

	
Virtually,

Bruce McFarling, Newcastle, NSW
ecbm@xxxxxxxxxxxxxxxxxxx