🚀 go-pugleaf

pet...@netcom.com (Loren Petrich) wrote:

>In article <32f5b72c.1563549@news>,
>Martin Gregorie <gregorie@see.sig.for.address> wrote:
>>nxh...@leonard.anu.edu.au (Nathan Hand) wrote:
>
>>>Those who do not know history are doomed to relive it... poorly.
>>As shown well by the computer industry:
>
>>- First there were mainframe O/S horrors such as OS/360 and its
>>predecessors, but some people did better and built George II, Multics,
>>and MCP
>
>	How was OS/360 so terrible?
>
OS/360 would have counted as OK if it had lived and died around 1965.
By 1970 it was distinctly archaic. We're talking OS360/MFT, not MVS. 

You had to know the contemporary oposition at the time. Multics was
past its prime and THAT could run 1000+ terrminals on a 24/365 basis
over multiple CPUs. George II/III was running well in 1970; I was
using it to run simultaneous batch & timeshare + RJE on a little 96Kb
mainframe with 24 Mb disk - OS/MFT couldn't match that.  VAX/VMS still
looks awfully like it owed a lot to George. You'll have to ask a
Burroughs person about MCP
 
>>- then minicomputers were invented and the state of o/s art reverted
>>to the likes of RSTS. Once again some people did better and came up
>>with UNIX (in 1970/71) and TOPS
>
>	How was RSTS so terrible?

Because when it came out it was only doing things in ways that were
already old hat on mainframes. Its authors shoudda read known their
mainframe history before writing it.

>
>>- then microcomputers were invented and  - you guessed it - the O/S of
>>choice turned out to be CP/M (based on RSTS, in 1978, yet!).
>>QDOS/MS-DOS is an obvious rip-off of CP/M rather than something more
>>advanced.
>
>	Actually, CP/M was based on RT-11 -- and QDOS/MS-DOS was in fact 
>a port of CP/M to the Intel-x86 architecture.

I rest my case. In 1981 the QDOS/MSDOS authors hadn't even read the
UNIX books. If they knew where mainframes & minis were at by 1980
they'd have done something MUCH better.

By the time QDOS/PCDOS/MSDOS were released in UNIX was old hat in
academic circles (it was running in 1970, remember) though the AT&T
sales policy kept it out of commercial use (free to universities, lots
of dosh and no support to anybody else). 

As an example of good contemporary micro operating systems in 1981
both Uniflex and OS-9 were running on 6809 boxes and were commercially
available. Both were modular UNIX clones with  shells, Unix-style
filing systems, and multi-user, multi-tasking capabilities. The 6809
was only an 8-bit machine but would easily support 3-4 users under
these systems. Uniflex is dead; OS9 is still very much alive.

>-- 
>Loren Petrich				Happiness is a fast Macintosh
>pet...@netcom.com			And a fast train
>My home page: http://www.webcom.com/petrich/home.html
>Mirrored at: ftp://ftp.netcom.com/pub/pe/petrich/home.html
>
>

----------------------------------------------------------------------
Martin Gregorie             |Logica UK Ltd
Greg...@logica.com         |All opinions expressed are solely those
+44 (0171) 637 9111         |of the author and not of Logica
----------------------------------------------------------------------