MARC (was Re: Thoughts on manual database design?)

Jay Jaeger cube1 at charter.net
Sun Sep 27 11:37:20 CDT 2015


On 9/27/2015 12:30 AM, Eric Christopherson wrote:

> On Tue, Sep 22, 2015, Jay Jaeger wrote:
>> (BTW, My memory of that acronym is "Machine Assisted Resource
>> Coordinator", a small-sized Unix work-alike developed by Ed Ziemba (RIP)
>> using Leor Zolman's BDS C compiler).
> 
> I'm having trouble finding much about this system; most of it is on your
> web page and the Wikipedia page for BDS C, which appears to borrow quite
> a bit from your page. Was MARC an OS itself, or a Unix-like layer on
> CP/M? Is it available to download and play with?
> 

I'm not surprised, as it was never available as a product.  I originally
inquired about it after an article about it in the BDS C User's Group
newsletter, and talked with Ed Ziemba, and he agreed to send me a copy
to play with / test.  I tested a few versions before Ed Ziemba's
passing, and did some work on programs like icheck/ncheck/dcheck.

It was an operating system, not a layer on top of CP/M, although it did
use the CP/M BIOS calling conventions for the I/O layer.  And, until you
got your own BIOS integrated into your copy of MARC, it could boot by
starting up a CP/M program on top of your existing BIOS, which was
termed a "parasitic boot".

I had a version of "ed" running on it, but the code was "tainted" as I
had access to a version done in PL/M that itself had ties to the
original Unix version, and had access to the Unix version code myself. I
also had a version of "make" than ran on it, that was not tainted.

(There is one article out there that supposes that MARC was doomed to
failure because the 8080 machine address was too small.  That is an
erroneous speculation from my perspective - given that programs like BDS
C, "ed" and "make" could run on it is pretty strong evidence to the
contrary).

Here is an article about the accident that pretty well sums things up:

https://books.google.com/books?id=pD0EAAAAMBAJ&pg=PT14&lpg=PT14&dq=marc+ed+ziemba&source=bl&ots=2kRendDrp1&sig=hilXBFd4S8tDb6Z9QFZMiNovLeo&hl=en&sa=X&ved=0CB0Q6AEwAGoVChMIn8XJ_c2XyAIVQQaSCh2qnwr4#v=onepage&q=marc%20ed%20ziemba&f=false

(Sorry about the fragmented URL).

As the above InfoWorld article mentions, Lauren Weinstein did work on
MARC after Ed's passing, but owing to legal difficulties (and perhaps
technical ones, too, I suppose), it never made it to market.  I
corresponded with Lauren back in the 80's as well as more recently when
I contemplated providing a copy to Al Kossow and the CHM.  Lauren
advised me against doing so because of the aforementioned legal
difficulties.  Lauren and Leor Zolman would know much more about all of
this than I did/do, as I never saw the source code.

I do not believe that the copy I had had the CP/M emulation component
that allowed CP/M programs to run on it that is mentioned in the article
above - I "just" had the Unix style system calls.

JRJ


More information about the cctech mailing list